(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Derailment of a CSX transportation freight train and fire involving butane, Akron, Ohio, February 26, 1989"

For Reference 



From the Library 

NATIONAL 
TRANSPORTATION 
SAFETY 
BOARD 



PB90-9 17006 
NTSB/HZM-90/02 



HAZARDOUS MATERIALS 
ACCIDENT REPORT 

DERAILMENT OF 

A CSX TRANSPORTATION FREIGHT TRAIN 

AND FIRE INVOLVING BUTANE 

AKRON, OHIO 

FEBRUARY 26, 1989 



For Reference 

Do Not Take 

From the Library 




The National Transportation Safety Board is an independent Federal agency 
dedicated to promoting aviation, railroad, highway, marine, pipeline, and 
hazardous materials safety. Established in 1967, the agency is mandated by the 
Independent Safety Board Act of 1974 to investigate transportation accidents, 
determine the probable cause of accidents, issue safety recommendations, study 
transportation safety issues, and evaluate the safety effectiveness of government 
agencies involved in transportation. 

The Safety Board makes public its actions and decisions through accident reports, 
safety studies, special investigation reports, safety recommendations, and statistical 
reviews. Copies of these documents may be purchased from the National Technical 
Information Service, 5285 Port Royal Road, Springfield, Virginia 22161. Details on 
available publications may be obtained by contacting: 

National Transportation Safety Board 
Public Inquiries Section, RE-52 
800 Independence Avenue, S.W. 
Washington, D.C. 20594 
(202)382-6735 



c 



L 



cs 



p\^ 



n 



5> 



TECHNICAL REPORT DOCUMENTATION PAGE 



1. Report No. 

NTSB/HZM-90/02 



2. Government Accession No. 

PB90-917006 



3. Recipient's Catalog No. 



4. Title and Subtitle Hazardous Materials Accident Report- 
Derailment of a CSX Transportation Freight Train and Fire 
Involving Butane, Akron, Ohio, February 26, 1989 



5. Report Date 

September 25, 1990 



6. Performing Organization 
Code 



7. Author(s) 



8. Performing Organization 
Report No. 



9. Performing Organization Name and Address 

National Transportation Safety Board 
Office of Research and Engineering 
Washington, DC. 20594 



10. Work Unit No. 

5092A 



11. Contractor Grant No. 



12. Sponsoring Agency Name and Address 

NATIONAL TRANSPORTATION SAFETY BOARD 
Washington, DC. 20594 



13. Type of Report and 
Period Covered 

Hazardous Materials 
Accident Report 
February 26, 1989 



14. Sponsoring Agency Code 



15. Supplementary Notes 



16. Abstract 

This report addresses the derailment of a freight train in Akron, Ohio, on February 26, 1989, and 
subsequent fire involving butane carried by the train. The safety issues discussed in the report are 
(a) the proximity of hazardous materials storage and plant facilities to mainline railroad tracks; (b) 
the lack of Federal requirements to maintain on board a train documents that identify the position 
and contents of cars carrying hazardous materials; (c) needs of emergency response personnel for 
technical assistance to evaluate dangers and risks during wreckage-clearing operations involving 
hazardous materials; (d) repair and inspection procedures at a rail car repair facility; (e) oversight 
of freight car repair facilities; (f) inspections of freight cars; and (g) maintenance and inspections of 
track. Safety recommendations were issued to CSX Transportation, Inc.; city of Akron, Ohio; 
Association of American Railroads; Federal Railroad Administration; National League of Cities; 
National Association of Counties; National Fire Protection Association; American National 
Standards Institute, Inc.; National Association of Regulatory Utility Commissioners; and the 
International Association of Fire Chiefs. 



17. Keywords 

hazardous materials storage; head shields; thermal protection; 
gib clearance; side bearing clearance; train consist; train 
profile; wreckage-clearing operations; car repair facilities; 
freight car inspections; track inspections; track maintenance 



18. Distribution Statement 

This document is available to the 
public through the National 
Technical Information Service, 
Springfield, Virginia 22161 



19. Security Classification 
(of this report) 

UNCLASSIFIED 



20. Security Classification 
(of this page) 

UNCLASSIFIED 



21. No. of Pages 

102 



22. Price 



NTSB Form 1765.2 (Rev. 5/88) 



CONTENTS 

EXECUTIVE SUMMARY v 

INVESTIGATION 1 

Events Preceding the Accident 1 

The Accident 4 

Witnesses' Observations and Actions 6 

Emergency Response and Wreckage Clearing 9 

Actions of Train Crewmembers 9 

Notification of Emergency 10 

Actions of Emergency Response Personnel 10 

Assessment of Conditions by Hazardous Materials Personnel 15 

Wreckage Clearing Operations 18 

Second Event Involving Tank Cars 21 

Emergency Response to the Second Event 22 

Injuries 25 

Damage 26 

Damage to Rai 1 Cars 26 

Damage to B.F. Goodrich Chemical Company 29 

Personnel Information 29 

Conductor 29 

Engineer 29 

Brakeman 29 

Fl agman 30 

Hazardous Materials Information 30 

Rail Car Information 30 

Car Repair and Rebuild Information 38 

Car Repair Shop of Northern Rail Car Corporation 38 

Shop Inspection Program of the AAR 39 

Car Rebuild Program of the AAR 40 

Shop Procedures of Northern Rail Car Corporation 

After the Derailment 41 

Other Car Repair Information 42 

Freight Car and Train Inspection Information 44 

Freight Car Inspections 44 

Inspections of Car WSOR 501003 45 

Tests of Train Air Brakes 46 

Track Information 47 

Track 47 

Description of Physical Evidence 48 

Track Inspections 52 

Track Maintenance 54 

Training and Emergency Preparedness 54 

Training of Railroad Personnel 54 

Emergency Preparedness 55 

Meteorological Information 56 

Toxicological Information 57 



m 



Tests and Research 57 

Event Recorders 57 

Tests of Locomotives 57 

Land-Use Information 58 

ANALYSIS 60 

General 60 

The Derailment 60 

Event Recorders 62 

Track Inspections 62 

Track Maintenance 63 

Freight Car Inspections 63 

Northern Rail Car Corporation Shop Procedures 64 

Association of American Railroads' Inspection of 

Freight Car Repair Shops 65 

Association of American Railroads' Inspection of Rebuilt Cars 65 

Tank Car Performance 65 

Initial Response to the Derailment by CSX Crewmembers and 

and Supervi sory Personnel 67 

Adequacy of Train Consist Information 73 

Movement of Damaged Tank Cars 74 

Location of Chemical Plants and Other Hazardous Materials Facilities 

Near Rai 1 road Tracks 76 

CONCLUSIONS 80 

Findings 80 

Probable Cause 81 

RECOMMENDATIONS 82 

APPENDIXES 87 



Investigation and Hearing 87 

Department of Transportation Emergency Response Guide 22 89 

Hazardous Materials Information 91 



IV 



EXECUTIVE SUMMARY 

On February 26, 1989, CSX Transportation, Inc., freight train 
No. D812-26 derailed at mile post 16.1 while traveling about 43 mph over 
Consolidated Rail Corporation (Conrail) main track No. 1, near the south end 
of Conrail 's rail yard, Akron, Ohio. Twenty-one freight cars in the train 
derailed, including nine tank cars filled with butane. The nine tank cars 
came to rest adjacent to a B.F. Goodrich Chemical Company plant, and butane 
released from two breached tank cars immediately caught fire. About 1,750 
residents were evacuated from a 1-square-mile area. On February 28, 1989, 
while some of the derailed tank cars were being moved from the accident site, 
one tank car full of butane rolled off its trucks; as a result, about 
25 families were evacuated from a second area. 

The National Transportation Safety Board determined that the probable 
cause of the derailment of train D812-26 in Akron, Ohio, on February 26, 1989 
was the inadequate rebuild and quality control procedures of the Northern 
Rail Car Corporation car repair facility and the inadequate inspections of 
car WSOR 501003 by designated car inspectors that permitted the car to enter 
and continue in service with excessive gib clearance and out-of-1 imits side 
bearing clearance. Contributing to the accident was the marginal condition 
of the track as a result of Conrail 's decision to delay rehabilitation of the 
track or not to place a slow order on the track. 

The following safety issues are discussed in this report: 

1. The proximity of hazardous materials storage and plant facilities 
to mainline railroad tracks; 

2. The lack of Federal requirements to maintain onboard a train a 
current consist to identify the position and contents of cars 
carrying hazardous materials; 

3. Local emergency response personnel needs for technical assistance 
to evaluate dangers and risks during wreckage clearing operations 
involving hazardous materials; 

4. Repair and inspection procedures at the rail car repair facility; 

5. Oversight of freight car repair facilities; 

6. Inspections of freight cars by designated car inspectors (carmen); 
and 

7. Maintenance and inspections of track. 

Recommendations concerning these issues were made to the CSX 
Transportation, Inc., the city of Akron, the Association of American 
Railroads, the Federal Railroad Administration, the International Association 
of Fire Chiefs, the National League of Cities, the National Association of 
Counties, the National Fire Protection Association, the American National 
Standards Institute, Inc., and the National Association of Regulatory Utility 
Commissioners. 



DERAILMENT OF A CSX TRANSPORTATION FREIGHT TRAIN 

AND FIRE INVOLVING BUTANE IN AKRON, OHIO, 

FEBRUARY 26, 1989 

INVESTIGATION 

Events Preceding the Accident 

On Sunday, February 26, 1989, about 4:00 p.m., the crewmembers for CSX 
Transportation, Inc. (CSX), freight train No. D812-26 reported for duty at 
the CSX train yard at Will ard, Ohio. The traincrew consisted of an engineer 
and brakeman in the front locomotive and a conductor and flagman in the 
caboose at the end of the train. All four train crewmembers were regularly 
assigned to this run; according to the CSX, the conductor was in charge. 
Train D812-26 was destined for the CSX Akron Junction train yard in Akron, 
Ohio, about 76 miles away (figure 1). The trip was a regularly scheduled 
run, operating round trip between Will ard and Akron, Sunday through Friday. 
On this day, the train was scheduled to make two stops to set off cars or to 
pick up cars en route to Akron Junction. 

Before leaving Will ard, the flagman picked up the train papers for the 
trip. Those papers included operating orders, a profile document, 1 and 
waybills. He gave a copy of those documents to the conductor and a copy of 
the profile and operating orders to the crewmembers in the front locomotive. 

Federal regulations require that a traincrew be provided a document that 
identifies the position in the train of each car carrying hazardous 
materials. 2 For this trip, the profile listed all freight cars in the train 
when it left Will ard, and the document identified the location of cars in the 
train that were carrying hazardous materials. The document did not identify 
the correct position of three cars in the train that did not carry hazardous 
materials; 49 CFR 174.26(b) did not require that those cars be listed on the 
document in their exact order. The document also did not list the caboose at 
the end. There was also no requirement that the document be kept current 
after the train left Will ard; Part 174.26(b) specifically relieves a carrier 
from the responsibility of amending the document to reflect the position of 
cars carrying hazardous materials in the train when the makeup of the train 



train's weight distribution) and list of cars in the train; special 
instructions that identified the proper shipping name and class of hazardous 
materials carried in the tank cars; a train switch list that listed the cars 
and identified locations where the cars were scheduled to be set off from the 
train; and emergency handling precautions for hazardous materials carried in 
the train. 

Title 49 CFR Part 174.26(b) states that "a train crew must have a 
document indicating the position in the train of each loaded placarded car 
containing hazardous materials, except when the position is changed or the 
placarded car is placed in the train by a member of the crew." 




Lake Erie 



t 
N 



i — ss= 1 



Map not drawn t 



Cleveland 



Willard 



Warwick ^ Akron 



Canton 



Figure l.--The scheduled route for train D812-26. 






is changed, or when cars carrying hazardous materials are added to the train 
after leaving the initial yard. 

According to the conductor, following his normal procedure, he observed 
the cars in the train as it departed the Willard yard and before he boarded 
the caboose. He later said that he checked the cars to determine (1) if the 
cars were grouped in the proper order for set-off during the trip, and (2) if 
cars containing hazardous materials were proper distances from the engine and 
caboose; 3 he did not take exception to the order of cars. He did not attempt 
to determine if all cars in the train were in the same order as listed on the 
profile. 

The profile document that was carried by the traincrew was generated 
from information entered into the CSX computer system. As the train departed 
the Willard yard, an employee in the yard tower observed the train and 
noticed that three cars in the train were in different positions than 
displayed on the computer list. He therefore changed the positions of the 
three cars on the computer; however, the new positions on the computer were 
not the same as the cars' positions in the train. 4 (When the derailment 
occurred later that evening, information retrieved by responding personnel 
about the makeup of the train reflected the order of cars as changed by the 
employee in the tower. An accurate list of the order of cars in the train 
when it left Willard was determined only after reviewing the location of cars 
at the accident scene.) 

When the train departed Willard at 5:10 p.m., it consisted of 

4 locomotives, 59 freight cars, and 1 caboose. Of the 59 freight cars, 
22 contained hazardous materials: 5 tank cars of chlorine, 1 tank car of 
potassium hydroxide, 15 tank cars of butane, and 1 tank car of butadiene. 

About 6:10 p.m., the train arrived in Easton, Ohio, the first scheduled 
stop. At Easton, three cars of lumber were set off by the brakeman; no cars 
were picked up at this location. The train then contained 56 freight cars. 
None of the crewmembers made notations on the profile document to identify 
the three cars set off from the train at Easton. Instead, the conductor 
followed company procedures and prepared a separate company document called a 
"wheel report" to identify those cars that were set off. 5 

About 6:37 p.m., the train arrived in Warwick, Ohio, the second 
scheduled stop. At Warwick, the brakeman set off 11 cars, including the 

5 tank cars of chlorine and the 1 tank car of potassium hydroxide. Also at 
Warwick, the brakeman added a caboose and four freight cars to the train; 
none of the cars picked up in Warwick contained hazardous materials. The 



Title 49 CFR Parts 174.85 through 174.93 contain safety requirements 
that address the placement of cars carrying hazardous materials in trains. 

This action did not change the order of hazardous materials tank cars 
on the list. 

The wheel report is not a document required by Federal regulations. 



train departed Warwick about 7:03 p.m. for Akron Junction with 49 freight 
cars and 2 cabooses. None of the crewmembers made notations on the profile 
document to identify the 11 cars set off at Warwick, or the 5 cars added to 
the train at Warwick. On the wheel report, the conductor identified the 
11 cars set off at Warwick, but he did not enter any information on the wheel 
report to identify the 5 cars added to the train at Warwick. The conductor 
later said he intended to enter that information on the wheel report after 
arriving in Akron. Figure 2 lists the position of all cars in the train when 
it left Warwick and identifies the cars that were set off and added en route. 

In addition to the documents that CSX required the conductor to carry 
with him, the conductor maintained a personal list of all cars involved in 
the train's movement. Whenever a car was set off or added to the train, he 
entered that information on his personal record. The conductor said he kept 
this list of cars "so that at any moment, if there is any question [to] 
arise, I can inform my head end [crew] exactly what we got in possible 
tonnage and so forth." 

The crew reported no problems or irregularities while operating the 
train over CSX eastbound track No. 2 from Willard to Warwick, about 60 miles. 
The train passed three equipment defect detectors that provide an audible 
warning to the locomotive engineer while traveling over this section of 
track; no defect warnings were received from two of the detectors, and the 
third detector reported a radio malfunction. 

The Accident 

At Warwick, the train switched to Consolidated Rail Corporation 
(Conrail) northbound track No. 1. The train entered the Conrail track at 
mile post (MP) 27.2 and passed two dragging equipment defect detectors: one 
at MP 23.0 and the second at MP 16.7. No defect warnings were received. 

The engineer reported that as the train approached MP 16.1, the train 
was "drifting" down a grade with the throttle in the idle position and with 
no brakes applied. He said the speed indicator showed that the train was 
traveling 42 mph. He stated further, however, that he believed the train was 
actually traveling about 40 mph--the maximum speed permitted on that section 
of track. He said that earlier in the trip he had checked the speed 
indicator on the lead locomotive unit (6124), and had calculated the actual 
speed of the train to have been about 2 mph slower than the speed displayed 
by the indicator. 6 

About 7:25 p.m., near MP 16.1 and the south end of Conrail 's rail yard, 
the locomotives crossed a bridge over the Ohio Canal and passed through two 
switching turnouts. 7 The first turnout connected the northbound main track 



A turnout is an area of track that contains a switch to direct a train 
from one track to another. 



Cars set off 
in Easton. 



Cars set off 
before accident 



f 1. ICG 978970 

2. TTPX 80315 

3. CPI 317106 





' 1. 


ECUX 


405280 




2. 


BO 


350077 




3. 


CO 


351022 




4. 


BO 


350977 


Cars set off 


5. 


ACFX 


87256 


in Warwick. t 


6. 


UTLX 


27831 Chlorine 


7. 


ACFX 


19628 Chlorine 




8. 


TLDX 


817067 Chlorine 




9. 


UTLX 


28180 Chlorine 




10. 


ACFX 


86995 Potassium 
Hydroxide 




i 11. 


ACFX 


86134 Chlorine 



/ 



y 



Order of 


cars when 








trs 


in le 


ft Warwick 








and 






at 


time of accident 




1. 


RBOX 


43456 






2. 


TTPX 


82333 






3. 


TTPX 


82745 






4. 


UP 


273208 






5. 


CO 


903042 


] 




6. 


PROX 


78942 


Cars 


added to train 


7. 


ACFX 


81549 


\ 


in Warwick 


8. 


NW 


178987 


(no hazardous materials). 


9. 

10. 
11. 


Nw 

SBD 
CGTX 


178526 


J 




128362 
23079 




— > 




12. 


WSOR 


501027 








13. 


WSQR 


501003 








14. 


WSOR 


501011 








15. 


MNPX 


2071 








16. 


PHD 


2129 








17. 


PHD 


2158 








18. 


CITX 


34602 


Butane 






19. 


UTLX 


88119 


Butane 






20. 


ZIPX 


3382 


Butane 




> Cars derailed 


21. 


CITX 


33875 


Butane 




in accident 


22. 


BCDX 


474 


Butane 




(cars 10-30 


23. 


GATX 


83935 


Butane 




inclusive) . 


24. 


UTLX 


804764 


Butane 






25. 


ACFX 


17285 


Butane 






26. 


CITX 


34944 


Butane 






27. 


WSOR 


501004 








28. 


WSOR 


501030 








29. 


NATX 


23894 








30. 


BO 
BO 


350777 




J 




31. 


356999 








32. 


PAL 


2193 






33. 


MRS 


2046 






34. 


PHD 


5015 






35. 


TILX 


300526 


Butane 




36. 


TILX 


300529 


Butane 




37. 


ACFX 


77100 


Butane 




38. 


TILX 


300460 


Butane 




39. 


CITX 


33831 


Butane 




40. 


BCDX 


404 


Butane 




41. 


MRS 


2090 






42. 


GONX 


310531 






43. 


GTW 


147637 






44. 


MP 


712197 






45. 


ACFX 


2406 


Butadiene 




46. 


CSTX 


138839 






47. 


CSTX 


136624 






48. 


GACX 


56174 






49. 


GACX 


56152 






50. 


SOU 


523330 






51. 


BO 


904063 


Caboose 







Figure 2. --Identification of cars involved in train movement. 



to the southbound main track, and the second turnout connected the 
northbound main track to a parallel auxiliary track. The engineer said that 
when the locomotives passed through the switches he noticed "a good bit of 
lateral movement. . .more lateral movement that Sunday night than there had 
been on the Friday [before] that I had worked." He said that after he 
noticed the lateral movement, which he believed to have occurred near the 
trailing point of the first crossover switch, he was going "to apply--use 
some air [brake] because of this lateral movement," but an emergency 
application of the train brakes was automatically initiated before he was 
able to do so. 

The brakeman, who rode in the lead locomotive with the engineer, said 
that the ride was rough at bridge 16 (over the Ohio Canal) and that they 
experienced lateral movement at that location every day. But, he added, "It 
didn't seem extreme." When asked specifically if there was more lateral 
movement on the day of the accident, the brakeman said, "Any time you have 
crossovers, there is always a little lateral movement. I did not take any 
exception to it... I don't believe it was much rougher than most of them." 

The engineer said that after the train brakes applied in emergency he 
immediately released the independent air brakes to prevent the brakes from 
being applied on the engine. The brakeman looked back and saw what he 
believed to be a tank car lying on its side in the southbound track. He 
then saw a shower of sparks, a big flash, and fire. 

At the time of the derailment, the train consisted of 4 locomotives and 
51 cars. Twenty-one cars derailed, including nine tank cars containing more 
than 270,000 total gallons of butane, a highly flammable liquefied petroleum 
gas. Cars number 10 through 30 derailed (see figure 2). 

Witnesses' Observations and Actions 

The accident occurred near a B.F. Goodrich Chemical Company plant, and 
several tank cars of butane came to rest adjacent to B.F. Goodrich's building 
number 315 (figure 3). One of these tank cars (UTLX 88119) lost all of its 
butane (about 30,000 gallons) after sustaining a tear 15 feet 9 inches long 
in its sidewall. A second tank car (CITX 33875) lost butane through a 
13-inch crack in its sidewall. A third tank car (ZIPX 3382) leaked butane 
through a valve in the dome. Tank car ZIPX 3382 was overturned and rested 
about 5 feet from the northwest corner of building 315; the tank car's 
pressure relief valve was obstructed by the ground and the car was involved 
in fire. Building 315 was 75 feet from the northbound track. 

Building 315 housed a B.F. Goodrich latex manufacturing operation. That 
operation involved the use of acrylonitrile, butadiene, and styrene monomer, 
which were stored in tanks nearby (see figure 3). Additionally, 




s- 

CD 



refrigeration equipment inside the building, used to control temperatures 
during the manufacturing process, contained anhydrous ammonia. 8 

A B.F. Goodrich employee was on the roof of building 332, located 
800 feet east of building 315, when he saw the train traveling past 
building 315. He said that when the train was opposite building 318, he saw 
a wall of fire develop on the north side of building 315 and that the flames 
were about three times as high as the four-story building. 

A second B.F. Goodrich employee, who was in the facility's power house 
on the west side of building 315, felt a strong vibration in the floor. He 
looked out a window and saw a bright light coming through it, so he went to a 
door on the east side of the building and stepped outside. He said that he 
saw a tank car moving through the air in the direction of the northwest 
corner of building 315; he thought that it then hit the building and came to 
rest directly beneath an overhead exhaust pipeline that carried unreacted 
styrene monomer fumes from building 315 to the power house. He then directed 
a shutdown and evacuation of the power house. He said that flames filled the 
area between building 315 and the power house. 

A third B.F. Goodrich employee, located in a control room on the third 
floor of building 315, heard a rumbling noise and looked to the north side of 
the building. Through a window he saw a wall of orange flames across the 
outside of the west half of the north wall. He stated that at first, he 
thought that the pipeline used to transport unreacted monomers to the power 
house had caught fire or exploded, but after checking gauges and 30 reactors 9 
he determined that was not the case. As a precaution, he closed a valve on a 
pipeline that supplied butadiene to the fourth floor and left the building. 
Once outside the building, he saw derailed tank cars involved in fire. 
Believing the tank cars could explode, he reentered building 315 and switched 
controls for the unreacted monomers to the emergency mode. 10 He then left 
the building, went to the storage tank area, and closed valves in pipelines 
that supplied acrylonitrile, styrene monomer, and butadiene to facility 
buildings. He returned to building 315, turned down reactor temperature 
controls, extinguished a north wall insulation fire 11 with a facility hose, 
and left the building. 



Characteristics of the chemicals mentioned in this paragraph are 
briefly described in the section, "Hazardous Materials Information." 



operation. 

1 

An action to slow the chemical reaction process. 

The fire was a result of butane burning and not from chemicals inside 
the building. 



Emergency Response and Wreckage Clearing 

Actions of Train Crewmembers .--When the train came to a stop, the 
flagman looked out the window of the caboose and saw a ball of fire over the 
B.F. Goodrich facility. He called the CSX Akron dispatcher on the radio, 
told him that they had a bad derailment, and asked him to stop trains in both 
directions. He told the dispatcher to call "haz mat" first, call the fire 
department, and then to call the Chessie [CSX] personnel because the accident 
was serious. 12 The flagman then walked past the rear of the train and set 
up signals to warn any approaching train to stop. 

The conductor said that he picked up his consist, 13 profile, and 
waybills, and stepped off the caboose. He walked toward the front of the 
train to survey the damage, and to determine which cars were involved in the 
accident. He marked on his copy of the profile the first car that he saw 
remaining on the rails at the rear end of the train, and after talking to the 
brakeman by radio, the conductor also noted on his copy of the profile the 
number and location of the last car that remained on the rails at the front 
end of the train. With this information, the conductor said that he 
determined that 21 cars had derailed and that 9 tank cars of butane were the 
only hazardous materials cars involved in the derailment. 14 The conductor 
then disengaged the locking mechanism between the couplers on the last 
derailed car and the following car, which was not derailed, so that cars not 
involved in the derailment could later be pulled away from the accident 
scene. He then returned to the back of the train where he and the flagman 
attempted to chase spectators away from the accident site. 

Immediately after the train stopped, the brakeman said that he stepped 
off the locomotive and walked ahead of the train to set up signals to warn 
any approaching train from the north to stop. He then returned to the 
locomotive, put on his coveralls, obtained the profile from the locomotive, 
and walked south toward the derailment to look for a fireman and to identify 
which cars were derailed. He found no fireman and decided that no one could 
get to the fire from that direction. He then disengaged a locking mechanism 
between the couplers of the eighth and ninth cars, and about 7:34 p.m., the 
engineer pulled the front end of the train forward about 400 feet. 



1 2 

The CSX Akron dispatcher stopped the movement of trains in both 

directions over that route and notified the Conrail dispatcher of the 
derailment. 



The list of cars in the profile document is referred to as 



13 
"consist" by train crewmembers and railroad personnel. 

Although the profile had not been revised to note that the tank cars 
of chlorine and potassium hydroxide had been set off before the derailment, 
the conductor knew that those cars had been set off before the derailment 
and he had noted the setoff of those cars on the wheel report. 



10 

About 7:53 p.m., the engineer, with the brakeman on board, pulled the 
front end of the train forward about 3/4 mile to an overpass at South Main 
Street. Afterward, they left the train and walked toward a fast food 
restaurant to get a soft drink when they saw a police car. They approached a 
police officer to make sure that emergency response personnel knew what was 
involved in the derailment and to identify themselves. About this time, the 
brakeman realized he had lost his copy of the profile. The police officer 
then drove the engineer and brakeman to see the fire chief. 

Notification of Emergency . --About 7:25 p.m., Akron police officers, 
firefighters, and residents observed a large fireball in the area of the 
South Akron rail yard and the B.F. Goodrich Chemical plant. Because of the 
fire's proximity to the plant, several persons reported that chemicals from 
the plant were burning, and by 7:30 p.m., the Akron Fire Department and the 
Akron Police Department dispatchers were inundated with calls from persons 
reporting the fire or seeking information about the emergency. 

A fire department lieutenant assigned to fire station 10, about 1 mile 
from the accident site, was one of the firefighters who saw the initial 
flash and fireball. He informed the battalion captain, who initiated an 
immediate fire department response to the B.F. Goodrich Chemical plant. 
Because the captain was aware that butadiene and other volatile chemicals 
were stored at the plant, he ordered all responding units not to proceed 
beyond the main gate of the B.F. Goodrich plant until the arrival of a chief 
officer from the Akron Fire Department. 

About 7:27 p.m., a hazardous materials coordinator from the Akron 
Police Department was assigned to coordinate an evacuation and to control 
traffic. A police command post was initially established about 1/3 mile west 
of the accident site, but it was later moved to the east side of the accident 
site where the fire department had established an incident command post. 

About 7:30 p.m., the CSX Akron dispatcher notified Conrail personnel at 
Warwick of the derailment and asked them to notify someone from Conrail at 
the South Akron yard. -About 7:33 p.m., the CSX Akron dispatcher called the 
Akron Police Department to report the derailment; the police stated they 
already knew about the accident. The CSX Akron dispatcher then notified the 
CSX headquarters operations in Jacksonville, Florida, and supervisory 
personnel in the Akron area. 

When a Conrail safety supervisor at home in Cleveland was notified about 
the accident, he instructed the Conrail Cleveland dispatcher to ask CSX to 
send a consist of the train to the Conrail Cleveland office. A consist was 
prepared by CSX personnel from the computer and was sent to Cleveland by 
facsimile equipment. The Conrail Cleveland safety supervisor then picked up 
the consist, dated February 26, 1989, 7:34 p.m., obtained self-contained 
breathing equipment, and drove to the accident site about 35 miles away. 

Actions of Emergency Response Personnel .--About 7:32 p.m., the district 
hazardous materials chief (operations chief) of the Akron Fire Department 
arrived at the B.F. Goodrich Chemical plant. B.F. Goodrich personnel met 
him at the gate and told him that they were not certain what had happened, 



11 

but they suspected a tank car, possibly filled with latex, 15 had erupted. 
However, because of the immense flames, smoke, and the tremendous noise-- 
which indicated to him that something was under pressure—the operations 
chief believed that the fire involved other products. The operations chief 
also knew that building 315 contained volatile, toxic substances because he 
had been at the plant several times during training to prepare for emergency 
situations. He initially established a command post near the entrance to the 
chemical plant and requested additional firefighters and equipment. The 
command post was later moved to a location about 3/4 mile east of the 
accident site. 

The operations chief said that his first concern, if tank cars were 
involved, was the possibility of a BLEVE, 16 and he directed the fire 
department dispatcher to contact Conrail for information about the contents 
of cars in a nearby rail yard. The operations chief could not see if tank 
cars were involved in the fire because building 315 was between him and the 
railroad tracks. Therefore, he asked the fire chief, who was responding to 
the accident site, to go to a location on the west side of the accident site 
where he believed the fire chief could see the tracks, could confirm if tank 
cars were involved in the fire, and could identify the product or products 
involved. The operations chief also ordered B.F. Goodrich personnel to 
activate four unmanned facility monitors, 17 which were permanently aimed at 
the chemical company's hazardous materials storage tanks, and to protect the 
tanks from thermal exposure in case the incident worsened. He also 
activated the city's mass evacuation plan. 18 

A few minutes later, the fire chief reported to the operations chief 
that tank cars were involved in the fire and that they were near the 
building; however, he could not read placards or markings on the tank cars to 
identify the product or products involved (figure 4). By this time, the 
mayor of Akron had also responded to the accident site where he met the fire 
chief. The mayor later told Safety Board investigators that his role was not 
to make specific, detailed judgments on what actions should be taken at the 
scene, but rather to make certain that the safest alternatives were chosen 
when key decisions affecting public safety were made; for example, decisions 



15 Finished product from the plant. 

16 Firefighting personnel often use the term BLEVE (boiling liquid 
expanding vapor explosion) to refer to the violent thermal rupture of a 
pressurized vessel that results in the instantaneous and violent release of 
burning gases. 

17 A monitor is a fixed or portable nozzle assembly used to direct a 
large volume of water for extinguishing fire or for cooling purposes, often 
referred to as a deluge. 

18 Areas adjacent to the accident site were evacuated immediately, and 
about 8:39 p.m. evacuation of a 1/2-mile radius was ordered based on the 1987 
Emergency Response Guidebook of the U.S. Department of Transportation. 



12 




Figure 4. --Derailed tank cars adjacent to building 315. 



concerning the size of areas that should be evacuated and the duration of 
evacuation. 

Between 7:44 p.m. and 7:50 p.m., the Akron fire department called the 
CSX Akron dispatcher and requested information about products carried in cars 
involved in the accident. The CSX dispatcher said that the train had a total 
of 14 tank cars of butane: 8 cars of butane near the head of the train 
followed by miscellaneous cars and then 6 more cars of butane. 19 Also, the 
dispatcher said that 1 tank car of butadiene was near the rear of the train. 
However, at 8:02 p.m., the Akron CSX dispatcher's office called CHEMTREC 20 
and reported that 9 tank cars of butane were involved in the derailment; he 
then gave to CHEMTREC the correct car numbers for those 9 cars. 



Actually, there were 15 tank cars of butane in the train: 9 near the 
front of the train and 6 that followed miscellaneous cars. 



Chemical Transportation Emergency Center operated by the Chemical 
Manufacturers Association. 



13 

About 8:17 p.m., the engineer and brakeman met with the fire chief in 
the fire chief's car. At that time, the brakeman said that butane was the 
only hazardous material transported in the derailed tank cars, that butadiene 
was not involved in the derailed cars, and that he had lost his copy of the 
profile. 

Because the fire chief wanted to see the paperwork to confirm that only 
butane was involved in the derailment, he drove the brakeman back to the 
accident site to look for the lost profile. When they did not find the 
profile, the brakeman contacted the conductor by radio to let the fire chief 
talk with him about the cars involved in the derailment. 

The conductor said he told the fire chief that according to his records, 
the only cars involved in the derailment that contained hazardous materials 
were nine tank cars of butane. During the conversation, the conductor knew 
he was talking to the fire chief, but the fire chief thought he was talking 
to someone at the rail yard. The fire chief was not aware, and neither the 
engineer nor the brakeman informed him, that a conductor and a flagman had 
been at the rear of the train and that the conductor had a second copy of the 
train profile. The conductor was aware that the brakeman had lost his copy 
of the profile, but he never informed the fire chief that he had a second 
copy at the rear of the train. Based on information from his conversations 
with the brakeman and the person on the radio (the conductor), the fire 
chief, at 8:38 p.m., told the operations chief by radio that nine cars of 
butane were involved in the derailment and that some of those cars were 
involved in fire. 

The CSX Akron Junction duty officer (trainmaster) arrived in the 
vicinity of the accident site about 8:20 p.m. and found the four locomotives 
and eight cars at the front of the train (which had not derailed) parked on a 
track near South Main Street; however, he saw no train crewmembers and could 
not contact them by radio at that time. About 8:35 p.m., the trainmaster was 
able to contact the flagman by radio. He asked if the conductor had given 
the fire department information about the hazardous materials that were 
involved in the derailment. The flagman responded yes, recalling the 
conductor's radio conversation with the fire chief and brakeman. As a 
result of his conversation with the flagman, however, the trainmaster 
incorrectly assumed that the conductor was with the fire chief and that the 
conductor had provided the fire chief all necessary information. The 
trainmaster later told the Safety Board that after his conversation with the 
flagman, he felt certain that the fire department knew what materials were 
involved and he decided to go to Akron Junction to make arrangements for the 
Federally mandated drug testing of the train crew, retrieval of speed tapes 
from the locomotives, and crew interviews. 21 



2 i 

A second CSX trainmaster arrived at a roadblock near the accident 

site about 8:30 p.m. He was directed to the fire chief and expressed his 

need to enter the accident site to assess what type of equipment would be 

needed for cleanup operations. He was denied entry and told to stay with an 

assistant fire chief until he was released about 10:30 p.m. to take train 

crewmembers to the hospital for drug testing. 



14 

About 8:41 p.m., the engineer took the four locomotives and front eight 
cars of train D812-26 (which had not derailed) to the CSX Akron Junction, 
about 4 miles north of the accident site, and the fire chief and brakeman 
returned to the incident command post from the accident site about 
8:55 p.m. 22 As had been instructed by the trainmaster, the conductor left 
the waybills for the derailed cars in the caboose at the accident site, and 
he took the train profile waybills for the non-derailed cars, emergency 
response guidance, and wheel report to the Akron Junction yard with him. 
Neither the conductor nor the flagman met with emergency response personnel 
before leaving the accident scene together by taxi to go to Akron Junction. 

Also about 8:55 p.m., the Conrail safety supervisor from Cleveland 
arrived at the accident site and said he was taken to the incident command 
post where the brakeman used the facsimile copy of the train consist brought 
by the supervisor to identify cars that had been set off before the accident, 
including all five tank cars of chlorine and one tank car of potassium 
hydroxide. The operations chief, however, was not convinced that only butane 
was involved in the derailment, and later told Safety Board investigators he 
did not remember that a train crewmember (the brakeman) had been present to 
help identify cars listed on the safety supervisor's copy of the consist that 
were involved in the derailment. Therefore, about 9:20 p.m. the operations 
chief said that he and the Conrail safety supervisor went to the accident 
site and compared the numbers on cars in the train to car numbers on the 
safety supervisor's copy of the consist. Even then the operations chief said 
he was not fully confident that the tank cars of chlorine had been set off 
before the accident. 

The CSX trainmaster received the train profile and wheel report from 
the conductor about 9:45 p.m., at the Akron Junction yard, and he had learned 
that the fire department was still concerned about the possibility of 
chlorine being involved in the derailment. He recognized the need to get the 
conductor's copy of the profile into the hands of the incident commander, 
drove to the incident command post, arriving at 10:15 p.m., and offered the 
incident commander the conductor's train papers. The incident commander 
(fire chief) referred him to the operations chief. The trainmaster explained 
to the operations chief, to whom he was directed, that the chlorine cars had 
been set off from the train before the accident, and that the butadiene 
car was not involved in the derailment. According to the operations chief, 
he then felt confident that chlorine was not involved in the derailment and 
decided he did not need the conductor's copy of the train papers. At that 



2 2 

The fire chief was the incident commander. Whenever it was necessary 

for him to leave the accident scene, the operations chief acted as incident 
commander . 



15 

time, the operations chief told the CSX trainmaster to pull the rear cars of 
the train (those that were not derailed) away from the accident site. 23 

Meanwhile, about 10:02 p.m., unmanned monitors were activated to direct 
water onto the tank cars involved in fire to cool them. The operations 
chief said that if the chlorine had been involved in the fire, he would not 
have used water to cool the cars because he believed that a runoff of 
chlorine could contaminate water supplies and produce chlorine gas. The 
operations chief also stated that the fire department probably could not have 
begun putting water on the fire any sooner than they did because the nearest 
water supplies they could use were 2,000 feet away. 

By 11:00 p.m., a 1 1/4-square-mile area had been evacuated involving 
about 1,750 persons 24 (figure 5). City schools and Akron University were 
closed, and city bus service was cancelled for the next day because the city 
bus garage was inside the evacuated area and not accessible to employees. 

Assessment of Conditions by Hazardous Materials Personnel .--About 11:30 
p.m. on February 26, the CSX division manager for hazardous materials (CSX 
hazardous materials supervisor) arrived at the accident site from Detroit, 
Michigan. He had previously met with and provided training to Akron Fire 
Department personnel and therefore was recognized by the operations chief. 
He stated that when he arrived, the issue of chlorine cars involved in the 
derailment was raised; he told the operations chief that all chlorine cars 
had been set off from the train before the accident. 

About 11:45 p.m., the CSX hazardous materials supervisor entered the 
derailment site to assess the damage to the tank cars. As a result of this 
inspection, he made the following observations: (1) tank car UTLX 88119 was 
probably in tact with no evidence of leakage; (2) flames were burning at one 
end of tank car CITX 33875; and (3) flames were burning near the dome of 
tank car ZIPX 3382, which was overturned; he estimated that one end of this 
car was located 4 to 10 feet from the northwest corner of building 315 
(figure 6). He took notes on the position of each tank car and later 
constructed a diagram of the car positions. While at the derailment site, 
the CSX hazardous materials supervisor also observed that firefighters were 
on the roof of building 315 fighting a roof fire, and that the ground fire 
from released butane had been extinguished. About 1:00 a.m. on February 27, 
two B.F. Goodrich employees were allowed to enter building 315 to check the 
cooling system that controlled reactions of butadiene and acrylonitrile in 
the latex manufacturing process. The cooling system functioned properly 
throughout the incident. According to the B.F. Goodrich plant manager, had 



23 



The rear cars of the train were pulled back to Warwick about 1:45 



a.m. the next morning, February 2 7, 1989. 



24 



Residents were not allowed to return to this area until 6:00 a.m. on 



February 28, 



16 




Figure 5. --Boundary of evacuation area ( £WW* ) . 
(Adapted from: Highway map of Summit County, Ohio, 1986. 
Summit County Engineers Office, Akron.) 



17 




^r- ° 



2 = 

> "3 



< ^ 



cn S- 

-i- 3 

u- -a 



18 

the system failed, chemicals involved in the manufacturing process could have 
heated naturally (by chemical reaction, polymerization), and vented toxic 
gases through the roof of the building. 

After returning to the command post, the CSX hazardous materials 
supervisor expressed his concern to the operations chief that the overturned 
tank car (ZIPX 3382) had a fire near the dome, and that the pressure relief 
device, equipped on top of the car, may not function because of the 
orientation of the car. Without proper functioning of the device, he said 
there was potential for a violent thermal rupture of the tank car. 

Meanwhile, a CSX superintendent of operations and the CSX Detroit 
division manager (in charge of the incident response for CSX) had arrived at 
the command post. These two men entered the derailment site with the CSX 
hazardous materials supervisor and a representative of the Association of 
American Railroads' (AAR) Bureau of Explosives. As a result of their 
inspection to assess the conditions, the CSX division manager told the fire 
department that it was critical to roll tank car ZIPX 3382 upright to lessen 
its potential for violent rupture. However, the CSX personnel insisted that 
Ashland Petroleum personnel be called to give an assessment. 25 

Wreckage Clearing Operations . --According to fire department personnel, 
throughout the incident, the city controlled the fire and activities at the 
scene, but it relied on the CSX for guidance and as the technical experts on 
wreckage clearing operations. The CSX division manager said that it was 
obvious the fire department had a lot of faith in the CSX hazardous materials 
supervisor because of his previous training with the city. (See section 
"Emergency Preparedness" for information on training.) The operations chief 
said that the city had no expertise to tell the railroad how to move the 
cars: "We had to leave that to their knowledge and ability." 

About 5:15 a.m, the fire department allowed CSX mechanical personnel to 
begin clearing a path to tank car ZIPX 3382 to prepare for rolling the car 
upright after the arrival of Ashland's personnel. According to the CSX 
division manager, each tank car was inspected and assessed to be safe to move 
by the CSX hazardous materials supervisor, the Bureau of Explosives 
representative, and a wreckmaster from either R.J. Corman Railroad 
Construction or Hulcher Service, Inc. (CSX-contracted wreckage clearing 
companies) before being moved. 

Ashland personnel from Ashland, Kentucky, arrived at the command post 
about 8:00 a.m. They completed an assessment of tank car ZIPX 3382 about 
10:00 a.m. and agreed that tank car ZIPX 3382 should be rolled upright. 



2 5 

The butane was en route to the Ashland Petroleum facility in Canton, 

Ohio. Ashland Petroleum was requested to send a team to the scene. The 
company initially told emergency response personnel that a team would be 
onsite at daybreak but then stated they would wait until morning and provide 
whatever assistance was needed. 



19 

According to the CSX general foreman from Willard, who oversaw the CSX 
mechanical wreckage clearing operations and who supervised the CSX 
wreckmaster, CSX mechanical personnel had assessed rerailing the tank cars, 
and they were concerned about the ability of damaged tank car body bolsters 
to support the weight of the tank cars. The general foreman also said that 
heavily damaged tank cars are very difficult to retruck (to place back onto 
wheel sets) after a derailment. He said that during early discussions of 
wreckage clearing plans for this accident, his first choice was to set the 
tank cars in a clearing to be off-loaded; he assessed the process for 
stabilizing tank cars on wheel sets as "dangerous--not the preferred method." 
He said he knew of no equipment available to determine if damaged body 
bolsters can provide adequate support on reassembled tank cars after the 
bolsters are involved in an accident. Instead, he said, personnel had to 
rely on visual inspections and experience. 

According to the CSX division manager, after assessments and discussions 
among CSX supervisory personnel, the AAR Bureau of Explosives representative, 
and the Ashland representatives, the group concluded that the safest action 
to take would be to transport the rerailed butane tank cars to a fixed 
facility for off-loading. The CSX general foreman said that the plan was to 
transport the tank cars to a fixed facility by loading the tank cars onto 
trucks (wheel sets) because flatcars (large enough to hold the weight of the 
tank cars) were not readily available. 

At that time, the Ashland representatives said that rerailed butane tank 
cars would be accepted at Ashland's fixed facility in Canton, Ohio, the 
cars' original destination; and the Norfolk Southern Railroad, the rail 
carrier serving the Ashland facility at Canton, agreed to accept and 
transport the tank cars. Therefore, CSX planned to rerail the nine tank 
cars, to move them about 4 miles to CSX's Akron Junction yard, and then to 
more permanently secure the seven tank cars still containing butane to their 
trucks for the 16-mile trip to Canton. 

The CSX division manager advised the fire department of the CSX plan to 
rerail the damaged tank cars and to move them to the Akron Junction yard; he 
did not explain any hazards that could be involved in moving the wrecked 
train, or provide any alternatives--such as loading the damaged tank cars 
onto flatcars instead of wheelsets. The operations chief told the Safety 
Board that he agreed with the CSX's plan to move the tank cars to the Akron 
Junction yard, and that he continued to rely on the railroad as the technical 
expert in these matters. According to both the fire department and the CSX, 
neither considered it safe to transfer butane between tank cars at the 
accident site because of the continuing fire from one tank car (CITX 33875) 
and the risk to the B.F. Goodrich chemical plant. The operations chief 
stated that he insisted that the cars not be unloaded at the accident site 
because of the continuing fire from tank car CITX 33875. (The car continued 
to burn until 3:00 p.m. on March 3, 5 days after the derailment.) 

About 3:25 p.m. on February 27, tank car ZIPX 3382 was rolled over to 
provide access to the dome cover. A butane leak was found at the packing 
beneath a valve attached under the dome cover. The valve handle was 



20 

tightened about 1/4 turn, the butane leak was stopped, and the butane fire at 
this location was extinguished. 

About 4:05 p.m., the CSX senior risk manager at CSX headquarters in 
Jacksonville, Florida, contacted the Federal Railroad Administration (FRA) of 
the U.S. Department of Transportation (DOT) in Washington, D.C., requesting 
permission to move the damaged tank cars as a hospital train 26 from the Akron 
Junction yard to Ashland's facility at Canton. The FRA granted permission on 
the condition that a buffer car be placed between each tank car. The CSX 
risk manager did not request permission to move the damaged cars from the 
accident site to the Akron Junction yard; he later said he considered that 
movement of cars a part of the emergency. 

About 8:55 p.m., tank car UTLX 88119, which was believed to be full of 
butane, was rolled upright. As the car was rolled, a coupler and attached 
drawbar dropped from the side of the tank; a penetrating gash 15 feet 9 
inches long (including a hole about 8 inches by 8 inches) was visible in the 
side of the tank car shell, which was found to be empty (figure 7). 

The CSX hazardous materials supervisor later told Safety Board 
investigators that he was not aware of any mechanical tests or examinations 
that were available to help assess the structural integrity of tank cars 
after a derailment. He said he depended on previous experience to evaluate 
the condition of derailed cars, and he acknowledged that because the full 
extent of damage to a tank car's integrity may not be visible, there will 
always be some risk when wreckage clearing operations involve tank cars of 
hazardous materials. 

By 3:00 a.m. February 28, seven of nine tank cars had been rerailed and 
were ready for transport to the Akron Junction yard. 27 CSX and Conrail 
representatives inspected the cars and agreed that they were safe to be 
moved. Three locomotives were connected first to two hopper cars serving as 
buffers and then placed in front of the seven tank cars. A wreck crane, a 
tool car, and a fourth locomotive followed. 28 Six of the tank cars were 
coupled together using chains because their couplers and/or drawbars were 
broken or missing. The tank cars had been chocked (shimmed) with wood 
between missing or damaged body bolsters and the trucks (wheelsets). The air 
brake equipment was damaged and was not functional. 



The term "hospital train" refers to cars that were damaged in an 
accident and that do not meet all FRA safety regulations. 

The remaining two tank cars were to be moved to Akron Junction at a 
later time. 



28 



The following units served as a barricade in the event that a tank 



car became free; the units also carried tools for use if needed. 



21 




Figure 7. --Tank car UTLX 88119. 



Second Event Involving Tank Cars . --About 4:00 a.m. on February 28, the 
wreck train began its movement to the Akron Junction yard. The plan was to 
move at about 2 to 3 mph, walking speed. Conrail and CSX employees rode 
on the train to observe the tank cars. As the engines began to pull the 
train, a chain between the first and second cars came loose; it was reapplied 
and adjusted. The engines again began to pull and the same chain broke. The 
cars were again chained together, and after moving about 1/2 mile, the chain 
between the first and second cars broke again. Another chain was then 
applied and the tank cars were inspected. Tank car GATX 83935 was found to 
be leaning, and when the train moved to take up slack, sparks were seen 
beneath the car. A wheel was found to be cutting the tank car steel jacket 
but it was not believed to have damaged the tank shell; therefore, jacks were 
used to raise the tank and additional wooden chocks were added between the 
tank and the truck. 



22 

As the train began to move a third time, two CSX employees and a Conrail 
employee walked beside the train to observe the tank cars. About 7:55 a.m., 
the train entered a long, superelevated curve; 29 about 1/4 mile into the 
curve (about 3 1/2 miles north of the original accident site), the crew 
observed tank car ZIPX 3382 rock and then roll off its trucks and onto the 
ground between the two mainline tracks (figure 8). A body bolster on the 
leading end of the car, supporting the tank car on the truck, had broken. A 
CSX supervisor within 30 feet of this tank car said there had been no warning 
of trouble. 

Emergency Response to the Second Event . --CSX and Akron fire department 
personnel checked the tank car for leaks using explosimeters; no leaks were 
found. Initially, the fire department wanted to evacuate a 1,500-foot radius 
from the site of the second event. The CSX hazardous materials supervisor, 
however, convinced the fire department to evacuate only the adjacent city 
blocks; four houses and two businesses were evacuated, requiring 35 persons 
to leave the area. The integrity of the tank car was assessed by CSX, Bureau 
of Explosives, and Ashland representatives. Following discussions between 
city officials and CSX personnel, the tank car was lifted back onto its 
trucks, and again chocked with wood between the tank shell and the trucks. 
The tank cars were moved to the Akron Junction yard, one or two at a time. 

The second event raised the city's concern about actions taken by the 
CSX to clear the wreckage. The mayor said that after the second event, his 
role changed out of necessity and that the city implemented "good, down-home 
common sense" rather than assume that the railroad was moving the cars by the 
safest possible method. He said that initially the city did not have a sense 
of authority to change plans affecting technical wreckage clearing 
activities, and that the city trusted and believed that someone with the 
proper expertise and knowledge was overseeing those operations. However, 
although continuing to rely on the railroad as the wreckage clearing experts 
after the second event, the city officials demanded an explanation of every 
detail and required several changes affecting cleanup activities. The mayor 
said that in spite of the second accident, the CSX personnel were 
cooperative throughout the clearing operations and he believed they were 
sincere in trying to safely clear the wreckage. 

The mayor said, however, that he was less than satisfied with the role 
of the FRA because agency personnel failed to notify the city of its presence 
onscene until after the second event. He believed that the FRA should have 
helped the city to determine if the damaged tank cars were capable of being 
safely moved from the accident site to the Akron Junction yard, 4 miles away, 
because the city had no expertise to determine if wreckage clearing 
activities proposed by the CSX were safe. 

Within an hour after the derailment, the FRA had dispatched four 
persons to the accident site to investigate the cause of the derailment, to 
observe actions taken, and to assure that those actions were in compliance 



29 



In a superelevated curve, one track is higher than the other to aid 



in controlling lateral forces on the train as it moves through the turn. 



23 




Figure 8. --Site of second event. 



24 

with the Federal regulations. The FRA's regional director stated, "It's the 
FRA's policy to not allow Federal personnel to offer instruction, advise or 
become involved in any advisory capacity to a railroad carrier, shipper, 
emergency personnel, including cities, during wrecking or clearing 
operations." The regional director also said, however, that if FRA employees 
saw a company preparing to lift a tank when they believed that action to be 
dangerous, they would discuss the matter with local officials. 

After the second event, the CSX division manager explained to city 
officials that there were three options for further handling of the damaged 
tank cars: (1) transfer the butane to other tank cars at Akron Junction, 
then move the trans-loaded cars to Canton; (2) move the damaged tank cars on 
to Canton as a wreck train; and (3) load the damaged tank cars onto flatcars 
for transport to Canton. The city officials opposed transferring the butane 
at the Akron Junction yard because doing so would expose a third area of the 
city to risks and would require another (precautionary) evacuation. The 
mayor also expressed concern about allowing the damaged tank cars to leave 
Akron because it would expose others to risk along the route. The CSX 
division manager recommended transferring the butane from the damaged tank 
cars at a fixed facility, where specific safety systems were installed that 
would minimize the risk of a major problem. Meanwhile, CSX personnel had 
located several 100-ton flatcars owned by the Department of Defense (DOD) and 
arranged for the flatcars to be transported to Akron Junction. CSX 
considered it safer to move the damaged tank cars to Canton on flatcars, 
rather than on wheel sets, and selected this option; Akron officials did not 
object. After determining that the tanks loaded on flatcars could be off- 
loaded at the facility, Ashland and the Norfolk Southern Railroad said they 
would accept the tanks on flatcars. 

Beginning March 2, CSX began loading the damaged tank cars onto the DOD 
flatcars at the Akron Junction yard for their movement to Canton. During the 
evening of March 3 and after several tank cars had been loaded onto 
flatcars, Ashland's vice president of operations told the CSX that it would 
not accept the damaged tank cars of butane at its Canton facility. 30 
Therefore, on the morning of March 4, CSX personnel met with the mayor, 
related Ashland's decision, and offered two alternatives: (1) to transfer the 
butane to other tank cars at Akron Junction; or (2) to move the damaged tank 
cars to a remote site near Warwick to transfer the butane. (Other transfer 
locations were considered and determined to be unacceptable by the Akron fire 
department because of nearby hazards or the lack of adequate water supplies.) 
After consulting with city officials in Warwick and arranging to provide 
firefighting support from Akron, the mayor of Akron agreed to transferring 
the butane at the remote site near Warwick; the tank cars were moved to the 
transfer point that evening. 

On March 5, about 6:30 a.m., CSX personnel and contract hazardous 
materials specialists met in Warwick to begin transfer operations; FRA 



The CSX division manager said that Ashland gave no reason for 
declining to accept the damaged tank cars. 



25 

personnel were present to determine if regulatory requirements were met. 
transfer operation began about 9:30 a.m. 



The 



On March 6, FRA personnel onscene advised the CSX that because of a 
severe wheel burn, tank car GATX 83935 could not be returned to Willard with 
the other tank cars after being emptied in Warwick. The CSX personnel 
disagreed, noting that inspections several days earlier revealed that the 
damaged area affected only the tank car jacket 31 and not the tank car shell, 
and that the full tank car had been permitted to move to Warwick. The FRA 
said that the car had moved to Warwick on an emergency basis (under 49 CFR 
174.47(b)) and that the movement was controlled and restricted (under 49 CFR 
215. 9). 3 * The FRA gave the CSX three choices for moving the car from 
Warwick: (1) Ashland must make a written request to the FRA in Washington, 
D.C., to move the car under a waiver; (2) CSX must make a written request to 
Washington, D.C., to move the car under a waiver; or (3) the empty car had to 
be purged. CSX elected to purge the empty car to complete the operations in 
a more timely manner. About 9:00 p.m. on March 6, 8 days after the 
derailment, all butane transfer operations were completed. 

Injuries 



Injury 33 

Fatal 

Serious 

Minor 



Railroad 
Personnel 







Emergency 
Personnel 



5 



Residents/ 
Passersbv 





50 



Total 





55 



TOTAL 



50 



55 



Of the 55 persons treated at five area hospitals, only 1 was admitted. 
He complained of difficulty breathing and chest pains; he had a history of a 
heart ailment and was admitted for observation. 

One police officer stationed at the perimeter of the evacuated area at 
the first derailment site complained of dizziness. He was treated for 
chemical burns on the hands and face, and acute bronchorrhea. No evidence 



The tank car jacket contains the thermal insulation surrounding the 
tank car shell. 

■z 2 

Title 4 9 CFR 215.9 allows defective freight cars to be moved to 

another location only for the purpose of effecting repairs, after it is 
inspected by an authorized person and determined to be safe to move the car. 
It allows a car to be moved to a location for unloading, if unloading the car 
is necessary for the safe repair of the car. If the car is empty, it may not 
be placed for loading. 



33 



Injuries are defined in 49 CFR 830.2 



26 

was found to indicate the source of the injuries, and no other emergency 
response personnel or residents reported similar symptoms. 

Four firefighters, who were on the roof of B.F. Goodrich building 315, 
were treated for possible exposure to ammonia fumes. They were treated and 
released. 

Fifty residents and passersby were treated for minor inhalation, 
coughing, and conjunctivitis. Some had eye irritation and most were treated 
for anxiety. 

Damage 

Damage to Rail Cars . --Of the 21 cars derailed in the accident, 16 cars 
were destroyed and 5 cars sustained moderate to heavy damage. CSX estimated 
the total damage to cars at $521,000. 

The trucks beneath the tank cars of butane were stripped from the cars, 
and some tank cars did not stay coupled together as a result of broken 
coupler shelves and broken drawbars. Several tank cars sustained impact 
damage to head shields, jackets, and tank shells. There were no tank car 
failures as a result of fire exposure. 

Four tank cars lost product as a result of derailment damage. The 
following is a summary of major damage to the nine derailed tank cars that 
transported butane: 

UTLX 88119 (DOT specification 112J340W). The tank shell sustained 
a penetrating gash 15 feet 9 inches long (including a hole in the 
tank shell 8 inches by 8 inches). (See figure 7.) When the tank 
was lifted, a drawbar and attached coupler were observed to drop 
from the side of the tank. The tank car jacket sustained localized 
areas of severe heat damage. Two body bolsters were bent 90°. The 
A-end coupler was broken and separated from the drawbar. The A-end 
stub sill sustained a 10-inch crack. 

ZIPX 3382 (DOT specification 112J340W). The tank car jacket 
displayed evidence of moderate burn damage, flame impingement, 
wrinkles and bulges from heat exposure, and tears. The top half of 
the B-end tank head sustained an impact that resulted in a dent, 
and about one-third of the head shield was missing (figure 9). The 
A-end coupler was missing and a body bolster at the B-end was 
missing. 

CITX 33875 (DOT specification 112J340W). The tank car shell 
sustained a 13-inch crack in the leading edge weld that connected 
the body bolster to the tank shell. Directly in front of the 
crack, the tank sustained an impact that resulted in a dent in the 
shell more than 4 inches deep. The width of the impact marks were 
about the same width as a coupler. The A-end top shelf coupler 
was broken, and the bottom shelf was bent downward and toward the „ 



27 




Figure 9. --Tank car ZIPX 3382, 



B-end of the car. The tank car jacket sustained tears and had 
evidence of flame impingement. The A-end head shield sustained an 
impact resulting in a dent about 5 inches across and two cracks, 7 
inches and 12 inches long. 

CITX 34602 (DOT specification 112J340W). An area about 8 feet 
long by 4 feet high was missing from the jacket on one side of the 
tank car. Some heat damage to the jacket was evident on the 
opposite side. The coupler shelves were broken on both ends of the 
car. 



BCDX 474 (DOT specification 105J300W). The A-end head shield 
sustained three impact dents, one about 4 feet in diameter in the 
top half (figure 10). The jacket on top of the tank sustained 
dents of about 2 feet and 1 foot in diameter. The tank car jacket 
sustained tears and wrinkles from heat exposure. 



28 




Figure 10. --Tank car BCDX 474. 



6ATX 83935 (DOT specification 112J340W). The jacket sustained an 
area of light burn damage on one side, heavy burn damage on the 
dome and B-end head, and tears. On the bottom side of the tank 
car, there was a 2-foot wheel cut through the jacket; a portion of 
the jacket 1/2 inch wide by 8 inches long was heat fused to the 
tank shell. Three body bolsters were bent and one was displaced. 
The coupler and drawbar on the A-end was missing, and the stub 
sill was deformed. 



UTLX 804764 (DOT specification 105J400W). The B-end coupler bottom 
shelf was broken off. The A-end stub sill was cracked. The jacket 
sustained impact dents and heat wrinkles from heat exposure. 

ACFX 17285 (DOT specification 112J340W). The A-end coupler was 
missing. The B-end top shelf coupler was missing. The jacket 
sustained a tear, dents, wrinkles from heat exposure, and a dent 
and gouges in the B-end head shield. 



29 

CITX 34944 (DOT specification 112J340W). Both couplers were 
missing. The A-end stub sill was broken. The jacket sustained 
tears and gouges, and damage from flame impingement. 

Damage to B.F. Goodrich Chemical Company . --B.F. Goodrich estimated total 
damage to the plant at about $1 million. Building 315 sustained fire damage 
from the butane fire to walls that faced the accident site, the roof, and two 
insulated pipelines. Moderate interior damage occurred, and electrical 
wiring that was subjected to heat had to be replaced. 

Personnel Information 

Conductor . --The conductor had been a railroad employee since June 26, 
1963. He was promoted to conductor on April 15, 1967. He had worked as a 
regular conductor on the Willard-to-Akron assignment since October 1988 and 
had previously held the job about 3 years earlier. 

The conductor had been off duty about 80 hours before reporting to 
work. He stated that he had slept about 9 hours Saturday night and had been 
awake about 5 hours before reporting for duty on Saturday. 

The conductor stated that he had had no emergency response training 
other than the operating rules class and reading the book of rules. He 
passed his most recent Operating Rules Examination on June 2, 1988, with a 
score of 96 percent correct. The conductor said that railroad management had 
checked his operational performance several times by observing his actions 
and by asking him questions; however, he had never been questioned about 
actions he should take in the event of an emergency. 

Engineer . --The engineer had been a railroad employee since August 13, 
1966. He was promoted to engineer on July 15, 1976, and had been assigned to 
the Willard-to-Akron run since April 1988. 

The engineer had been off duty 37 1/2 hours before reporting for work on 
the day of the accident. He said that he had slept about 7 hours the night 
before the accident, awakening about 8:00 a.m. on the day of the accident. 

The engineer said that he had not been provided any specialized training 
for emergency response actions when hazardous materials were involved in 
accidents, and that he did not recall any operating rules test questions on 
the subject. He passed his most recent Operating Rules Examination on 
September 29, 1988, with a score of 88 percent correct. 

Brakeman .--The brakeman had been a railroad employee since March 30, 
1968. He was promoted to conductor on September 16, 1982, and had been 
assigned to the Willard-to-Akron run regularly for several months. 

The brakeman had been off duty 37 1/2 hours before reporting for work on 
the day of the accident. He said that he had slept about 8 hours Saturday 
night and that he had been awake about 6 hours before reporting for duty on 
Sunday. 



30 

The brakeman stated that he had received no special training about what 
actions he should take when there was an accident involving hazardous 
materials; however, he said he knew the importance of locating emergency 
response personnel. He passed his most recent Operating Rules Examination on 
June 2, 1988, with a score of 94 percent correct. 

Flagman . --The flagman had been a railroad employee since July 23, 1965. 
He was promoted to conductor on January 1, 1969, and had been assigned to 
the Willard-to-Akron run on a regular basis. 

The flagman had been off duty 37 1/2 hours before reporting for work on 
the day of the accident. He said that he had slept from 8 to 10 hours 
Saturday night and that he had awakened at 9:00 a.m. on Sunday, the day of 
the accident. 

The flagman stated that he had not been provided any specific emergency 
response training, and that he was not familiar with instructions in the 
timetable or operating rules that addressed what should be done with train 
papers after an accident. He passed his most recent Operating Rules 
Examination on June 9, 1988, with a score of 96 percent correct. 

Hazardous Materials Information 

The train carried about 450,000 gallons of butane: the 15 tank cars 
contained about 30,000 gallons of butane each. Nine of those cars (carrying 
about 270,000 gallons of butane) were derailed. An estimated 70,000 gallons 
of butane were consumed in the postaccident fire. 

Butane is a liquefied petroleum gas and is classified by the DOT as a 
flammable gas. It has a flashpoint of about -100 °F, has flammable limits in 
air of between 1.8 percent and 8.4 percent. Although butane is not toxic, 
exposure to its vapors may cause dizziness or suffocation. Initial emergency 
response actions for accidents involving butane are contained in the DOT's 
1987 Emergency Response Guidebook, Guide 22. (See appendix B.) 

Three hazardous materials used in the chemical plant's manufacturing 
process were present in B.F. Goodrich Chemical Company's building 315 and 
were stored in tanks nearby: (1) butadiene, (2) styrene monomer (toxic, 
flammable liquid); and (3) acrylonitrile (flammable liquid and poisonous). A 
fourth hazardous material, anhydrous ammonia, a nonflammable compressed gas 
that is extremely corrosive to human tissue, was also present in the plant's 
refrigeration equipment used to control chemical reactions in the 
manufacturing process. Detailed information on the hazard characteristics of 
these materials is contained in excerpts from the U.S. Coast Guard's Chemical 
Hazardous Response Information System (CHRIS) Manual. (See appendix C.) 

Rail Car Information 

A postaccident inspection of the 21 cars involved in the derailment 
resulted in the identification of one car, WSOR 501003, with preaccident 
mechanical deficiencies. It was the 13th car in the train behind the 



31 

locomotives (cars in positions 10 through 30 were derailed). No evidence of 
prederailment mechanical deficiencies were found in components of the other 
cars. 

Freight car WSOR 501003 was a 100-ton, two compartment covered hopper 
car (figure 11). It had a maximum cargo limit of 204,580 pounds, a light 
weight of 58,420 pounds, and was transporting about 197,580 pounds of sand at 
the time of the accident. It was originally built in October 1965 and was 
rebuilt in October 1988. After being rebuilt, car WSOR 501003 had made two 
successful round trips before the trip involving this accident. 

After the derailment near the B.F. Goodrich plant, WSOR 501003 was 
observed turned over on its left side (direction of movement of the train). 
The A-end truck was separated from the car and was found about 50 feet south 
of the derailed car. The B-end truck, which was the lead truck in the 
direction of movement of the train, was also separated from and in line with 
the car. Heavy rail burns were observed on the left sideframe of the B-end 
truck and wheel marks were observed on the left side cross-ridge slope sheet 
of the car (figure 12). Additionally, the lead wheelset showed evidence of 
heavy chaffing and bluish discoloration on the steel rims and flanges. Some 
of the springs were dislodged from the right sideframe of the B-end truck 
assembly. Figure 13 identifies some pertinent components of a truck 
assembly, including the bolster and sideframes. 

After WSOR 501003 was rolled upright on its own trucks, the truck 
bolster on the A-end of the car was observed to be a different design from 
the truck bolster on the B-end of the car, and that the truck bolster pattern 
numbers were different. The A-end pattern number was Scull in AAR-B-1925 and 
the B-end pattern number was Gould-B-9S-14EJ-BX. 




Figure 11. --Freight car WSOR 501003. 



32 





Figure 12. --Some of the damage observed on car WSOR 501003: 
(A) rail burn on sideframe, (B) wheel marks on slope sheet. 



33 




34 

Rule 47(B)(1)(a) of the Field Manual of the AAR Interchange Rules 
adopted by the AAR's Mechanical Division, Operations and Maintenance 
Department, states that truck bolsters must not be used if they do not have 
AAR identification marks or pattern numbers. Rule 47(B)(2) states that 
truck bolsters arranged for "wide land" bolster rotation stops are not 
interchangeable with bolsters incorporating the standard rotation lug. The 
B-end truck bolster on car WSOR 501003 (Gould-B-9S-14EJ-BX) was a wide land 
design, and it did not have AAR identification marks or pattern numbers. 34 

The gib clearance of the A-end truck was noticeably different from the 
clearance of the B-end truck. (A gib clearance is the space between the 
truck bolster gibs and the sideframes; it allows the truck components to 
guide the truck through the contour of the rail as the wheels of a car follow 
the track.) Total lateral gib clearance of the A-end truck was a maximum of 
1 inch at each location measured; total lateral gib clearance of the B-end 
truck exceeded 3 inches at each location measured. Figure 14 illustrates 
the specific gib clearance measurements recorded for the B-end truck. 
Figure 15 shows the gib clearance visible outside the sideframe; the 
additional gib clearance inside the sideframe is not visible in the 
photograph. 

Rule 47(A) of the Field Manual of the AAR Interchange Rules provides 
maintenance and gauging standards for truck bolsters. The rule states that 
when wheels are changed or trucks are dismantled, bolster gibs must be 
measured before disassembly, and when wear exceeds limits of 1 1/2 inches 
between the bolster and truck sideframe (for the type equipment installed on 
WSOR 501003), the equipment must be repaired to maximum nominal clearances 
between the bolster and truck sideframes of 1/2 inch inside and 1/2 inch 
outside. The trucks on both ends of WSOR 501003 had been dismantled when the 
car was rebuilt. The FRA has no safety regulations that address gib 
clearances between bolsters and truck sideframes. 

The B-end truck bolster had originally been designed for use with 
friction journal bearings; however, letters "RB" cast in the truck bolster 
(separate from the pattern number) indicated that the bolster had been 
modified to accept journal roller bearings. The sideframes on the B-end and 
the A-end of the car were standard, and built for 6 1/2-inch by 12-inch 
journal roller bearings. The A-end and B-end trucks had been manufactured by 
ASF Ride Control, were equipped with seven coil cluster D-3 springs, and were 
set on 36-inch wheels. 



There are two designs of bolsters: wide land and standard rotation 
lug. Each design requires a certain sideframe; a bolster of one design is 
not to be used with the sideframe for the other design. The B-end truck on 
car WSOR 501003 had sideframes designed for a standard rotation lug bolster 
instead of sideframes designed for a wide land bolster. 



35 



truck bolster gib 



d 



left sideframe 



_/-b-.. 



g 



right sideframe 



a = 2 inches e = 1 7/8 inches 

b = 1 1/4 inches f ■= 1 1/4 inches 

a + b = 3 1/4 inches e + f = 3 1/8 inches 



c = 2 inches g = 1 3/4 inches 

d = 1 3/16 inches h = 1 3/8 inches 

c + d ■= 3 3/16 inches g + h ■= 3 1/8 inches 



Figure 14. --Gib clearances measured between truck bolster gibs and sideframes 
on the B-end of WSOR 501003. 







Figure 15. --Gib clearance visible outside the sideframe. 



36 

After car WSOR 501003 was rerailed and after its load of sand was 

leveled, the side bearing clearances were measured. (A side bearing 
clearance is the distance between a side bearing roller on a truck and a side 

bearing plate attached to the car body bolster above the roller; this 

clearance is illustrated in figure 16.) (One of the two truck side bearing 

rollers on the A-end right side was missing.) The measurements were as 
follows: 



B-end Truck 



Left Side (BL) 11/16 inch 

Right Side (BR) 4/16 inch 

Total (B-end) 15/16 inch 35 

A-end Truck 

Left Side (AL) 8/16 inch 

Right Side (AR) 2/16 inch 

Total (A-end) 10/16 inch 

Diagonal Computations 

BL plus AR 13/16 inch 35 

BR plus AL 12/16 inch 



FRA safety regulations (49 CFR 215.119(c)) prohibit a railroad from 
placing into service, or continuing in service, a car if any of the 
following conditions exist: 

(1) Part of the side bearing assembly is missing or broken; 

(2) The bearings at one end of the car, on both sides, are in contact 
with the body bolster (except by design); 

(3) The bearings at one end of the car have a total clearance from 
the body bolster of more than 3/4 (12/16) of an inch; and 

(4) At diagonally opposite sides of the car, the bearings have a total 
clearance from the body of more than 3/4 (12/16) of an inch. 



There is no minimum clearance; however, constant contact is not permitted for 
the type of bearing used on WSOR 501003. 



37 




HHM 



Figure 16. --Components of a side bearing: (A) the side bearing 
plate attached to the car body bolster, (B) the roller attached to 
the truck, and (C) the side bearing clearance in-between. 



The Field Manual of the Interchange Rules of the AAR, Rule 47(E)(2), 
however, states that side bearings must have 3/16-inch minimum to 
5/16-maximum clearance. (This clearance does not apply to constant-contact 
type side bearings.) The AAR considers the FRA regulations to be safety 
standards, and it considers the AAR rules to be maintenance standards that 
are designed to identify and correct problems before the FRA regulations are 
violated. The AAR considers the FRA maximum side bearing clearance to be 
safe. 



Further visual inspection of car WSOR 501003 disclosed previous cracking 
in the A-end, right side, body bolster diaphragm where it had been welded to 
the center sill. The cracks were progressing toward the outer edge of the 
car. A mixture of common and high tensile bolts secured the body center 



38 

plate to the car. 36 Additionally, the body bolster wear plates on the B-end 
of the car displayed evidence of pounding or striking. 

Car Repair and Rebuild Information 

Car Repair Shop of Northern Rail Car Corporation . --The WSOR 501003 
covered hopper car was one of 36 similar 100-ton hopper cars rebuilt several 
months before the accident at a contract car repair shop owned by Northern 
Rail Car Corporation (NRCC). In addition to WSOR 501003, another 4 of the 36 
rebuilt hopper cars were involved in the derailment. The 36 rebuilt hopper 
cars were identified as car numbers WSOR 501001 through 501036. 

The 36 covered hopper cars had been purchased by the Northern Rail Car 
Leasing Company of Cudahay, Wisconsin, in April and June 1988. The cars were 
taken to Cudahay, a suburb of Milwaukee, where they were rebuilt at the NRCC 
freight car repair facility. It offers general freight car repairs, major 
wreck repairs, and locomotive repairs as a contract repair shop. 
Additionally, the company operates a wheel shop, designs and builds special 
equipment cars, and offers a mobile repair service. The repair facility has 
been under its current ownership since April 1986. 

The car repair shop estimated that about half of its work in 1988 was 
performed for the Wisconsin & Southern Railroad Company of Horicon, 
Wisconsin, and the Northern Rail Car Leasing Company of Cudahay. The 
Wisconsin & Southern Railroad Company operates as a railroad over 147 miles 
of track between Northern Milwaukee and Oshkosh, Wisconsin. In February 
1989, it was operating 3 locomotives, had about 350 cars, and had been under 
its current ownership since August 1988. The three companies--NRCC, the 
Wisconsin & Southern Railroad Company, and the Northern Rail Car Leasing 
Company- -were owned by the same person. 

Between March 3, 1989, and April 6, 1989 (after the accident), CSX 
inspected 24 of the 36 hopper cars rebuilt by NRCC found moving on the CSX 
system but that had not been involved in the accident on February 26; special 
attention was given to side bearing condition and clearance, and truck 
bolster lateral gib clearance. Of the 24 cars, 3 had side bearing 
clearances that exceeded FRA limits for diagonally opposite ends, and 1 car 
had no side bearing clearance at the left or right side on one end of the 
car. (None of the 24 cars was equipped with constant-contact type bearings.) 
Additionally, 1 of the 24 cars had a cracked side bearing top plate, 4 cars 
had loose side bearing top plates, and 1 car had a 7-inch crack in a body 
bolster web. 

According to NRCC records, 32 of the 36 cars rebuilt at NRCC had side 
bearing repairs made during the rebuilding process. These repairs consisted 
mainly of roller renewal, roller housing repair, and the addition of shims to 
the trucks or car bodies. 



36 



Rule 60(B)(2) of the AAR's Field Manual of the Interchange Rules 



requires the body center plate to be secured with high tensile steel bolts 



39 

During a postaccident inspection of WSOR 501006 at the Wisconsin & 
Southern Railroad facility in Horicon on April 11, 1989, Safety Board 
investigators found that the truck sideframes on the A-end of the car did not 
conform to the requirements of the AAR Manual of Standards and Recommended 
Practices described in the section "Method of Marking and Mating Cast Steel 
Truck Side Frames." The recommended standard establishes: 

...a uniform practice for marking truck side frames to permit 
pairing of frames on the same truck, with respect to dimensions 

between centers of journal boxes (wheelbase) Under the adopted 

system, the number of buttons remaining on the side frame indicates 
the variation over the nominal dimension, by tabulation, and frames 
having the same number of buttons or within one of the same number 
are selected for use on the same truck. By following this method, 
the variation in wheelbase dimension would not exceed .150 inch. 

The A-end left truck sideframe had four buttons, and the A-end right truck 
side frame had two buttons. 

Shop Inspection Program of the AAR . --Rule 120 of the Field Manual of the 
AAR Interchange Rules authorizes the AAR's Mechanical Inspection Department 
(MID) to conduct inspections and investigations to check compliance with the 
Interchange Rules and the mandatory provisions of the AAR's Manual of 
Standards and Recommended Practices. MID inspectors are permitted to inspect 
freight car component repair and reclamation facilities and to analyze car 
repair records to detect uncommon trends in the repair of cars and to 
identify problems with specific components, rules, or billing charges. The 
rule authorizes the inspection of signatories to the Interchange Rules; 
railroads and private car companies not signatories to the rules; contract 
shops engaged in building, rebuilding, repairing, or demolishing cars; and 
plants engaged in the manufacture, repair, and reclamation of material. 

Section D of Rule 120 authorizes a $750 charge for certification of a 
nonmember's facility to defray the expenses of an inspection, and the rule 
states that if a certified shop is temporarily shut down for more than 12 
months, the certification will be withdrawn. AAR representatives stated, 
however, that the AAR neither certifies nor approves contract repair shops. 
Instead, the referenced certifications are for wheel shops; roller bearing 
reconditioning shops; and air brake, coupler, and draft gear reconditioning 
facilities because of the precise work required at those facilities. The AAR 
has not believed it necessary to certify freight car repair ("rip track") 
shops. 

According to AAR personnel, although a contract freight car repair shop 
is not required by the AAR to seek an inspection of its facility by an MID 
inspector, railroads providing service to a contract repair shop ask the 
facility, as a practice, to obtain an MID inspection as an assurance that the 
shop is providing an acceptable quality of work. During a routine inspection 
of a contract repair shop, an MID inspector determines if proper equipment 
and manuals are available to do repair work offered, if repairs are made 
according to interchange rules and standards, and if there is an adequate 
recordkeeping system. Additionally, MID inspectors examine one or more 



40 

repaired cars leaving the shop, if available, to determine if the cars meet 
all mechanical requirements and if any defects have been overlooked. 

The owner of the NRCC contract car repair shop testified that it was his 
understanding that the AAR certified rebuild shops, or the rip track portion 
of the shop, after an inspection of the facility. He said he believed the 
inspections were necessary because private car owners and railroads 
considering use of the facility want to know if the shop is AAR-approved. 

A routine inspection of the NRCC contract car repair shop was conducted 
by an AAR MID inspector on January 14-15, 1987. The MID inspector noted the 
following problems: (a) operative roller bearing grease gun not available; 
(b) AAR M-963 journal box lubricating oil not available; (c) center plate 
lube not available; and (d) steeple back journal bearing wedge condemning 
gauge not available. No repaired cars leaving the shop were inspected 
because none were available at that time. 

An officer of the repair shop was present during the inspection and was 
provided a copy of the inspection report. In a letter dated February 6, 
1987, to the president of the repair shop, the AAR asked that exceptions be 
corrected because they were serious and asked to be advised when the 
corrections were made. A copy of the inspector's report was sent to the 
railroad that provided service to the repair shop, in accordance with normal 
procedures, to alert the railroad of deficiencies found. A railroad may then 
adjust the level of inspection activities as necessary to determine the 
mechanical condition of cars the railroad would pick up at the shop. The AAR 
had no records to confirm that the exceptions were corrected, and no followup 
routine inspection was performed as of June 1989. 

The AAR stated that although they desire to conduct followup routine 
inspections of car repair shops every 12 to 18 months, limited staff 
resources and an increasing number of car repair shops do not permit followup 
inspections within that time. The AAR has 10 MID inspectors qualified to 
conduct routine inspections — including 4 wheel shop inspectors — and about 
2,000 shops that require inspection. Of the 2,000 shops, the AAR estimates 
that 800 shops are contract car repair shops and that 200 are reconditioning 
facilities for wheels, axles, and other freight car components. The 
remaining 1,000 are primarily railroad (carrier)-operated shops. 

The FRA has no regulations governing the operation of freight car repair 
shops, and therefore does not inspect such shops. Rather, the FRA requires 
that all freight cars meet or exceed minimum FRA safety regulations when 
transported. 

Car Rebuild Program of the AAR . --Freight cars rebuilt in accordance with 
AAR procedures are given new "rebuilt" birthdays, and they are assigned 
higher book values that reflect expenses paid by the car owners to improve 
the cars. To qualify for AAR rebuilt freight car status, cars must be 
rebuilt to all current AAR Interchange Rules and FRA safety regulations, must 
be returned to the same type service (for example, a hopper car must remain a 
hopper car), and must be more than 10 years old but less than 25 years old; 
appropriate paperwork must also be completed and approved by the AAR. 



41 

After purchasing the 36 20-year-old hopper cars, Northern Rail Car 
Leasing notified the AAR by telephone of its intent to rebuild the hopper 
cars and then sent the required forms and fees to the AAR to begin the 
qualification process. On July 18, 1988, the Northern Rail Car Corporation 
notified the AAR that a sample (representative) rebuilt car was available for 
inspection. The AAR's procedures require that one or more rebuilt cars in a 
project be inspected by an MID inspector to determine if the car(s) meet all 
current AAR Interchange Rules and FRA safety regulations. 

On July 19, 1988, an MID inspector arrived at the NRCC facility and 
inspected one of the rebuilt cars from the project, car WSOR 501032. He 
completed an inspection form, noting that the car met all requirements except 
being weighed. The inspector told NRCC personnel to rebuild the remaining 
cars in the same manner. The MID inspector later said that he inspected only 
one rebuilt car because it was the only one of the 36 available. The owner 
of the facility said, however, that he believes five rebuilt cars were 
available because the cars were numbered as they were completed, from the 
highest number to the lowest, and therefore cars WSOR 501036, WSOR 501035, 
WSOR 501034, and WSOR 501033 should have also been finished. 

On September 26, 1988, the AAR notified the NRCC by letter that WSOR 
501032, inspected on July 19, 1988, met all mechanical requirements and had 
been approved by the AAR's Mechanical Division, and that approval forms had 
been forwarded to the AAR's Transportation Division for approval and 
handling. The AAR letter also requested another form from NRCC, listing 
costs associated with rebuilding the cars, so that official rebuild status 
and new car values could be assigned. The NRCC submitted the additional form 
to the AAR on December 6, 1988. At the time of the Akron accident, the AAR 
had not yet approved official rebuild status of the hopper cars. 

On November 21, 1988, another MID inspector, while inspecting cars for 
proper cargo loading at a railroad yard in Chicago, Illinois, inspected WSOR 
501017, one of the 36 rebuilt cars. He noted that the car was marked with a 
recent rebuilt date. He saw that the car did not comply with AAR rules 
because it was equipped with threaded fittings in the air brake line (Rule 
4(B)(2) of the Field Manual of the AAR Interchange Rules) and that it did not 
comply with the FRA safety regulations (49 CFR Part 231, Railroad Safety 
Appliance Standards) because it was equipped with side ladders extending from 
the bottom to the top of the car. Because all 36 rebuilt cars were equipped 
in a similar manner, a telephone agreement was reached between the Wisconsin 
& Southern Railroad and the AAR requiring the Wisconsin & Southern Railroad 
to correct these deficiencies as the cars returned to the railroad's terminal 
in Horicon. During the Safety Board's visit to the Wisconsin & Southern 
facility at Horicon on April 11, 1989, Safety Board investigators determined 
that as of April 11, 1989, these deficiencies had been corrected on 25 of the 
36 cars. According to the owner of the Wisconsin & Southern Railroad, all 
36 cars were repaired by about May 26, 1989. 

Shop Procedures of Northern Rail Car Corporation After the Derailment . -- 
A Northern Rail Car Corporation repair shop manager, who was a car repairman 
at the time of the WSOR hopper car rebuild project, said that quality control 
procedures had been improved since the Akron derailment. He stated that a 



42 

quality control inspection sheet was now used in conjunction with the repair 
of all cars. Among other requirements, the following items must now be 
checked before repaired cars are approved by the NRCC supervisors: (a) the 
measurement of gib clearances; (b) the measurement of side bearing 
clearances; and (c) the proper match of buttons on sideframes. 

He also stated that more frequent inspections are made by supervisors 
while cars are being repaired and that two supervisors must inspect finished 
cars, using a checklist, to be certain that all repairs had been performed 
properly and that needed repairs were not missed. 

The NRCC shop manager told the Safety Board that excessive gib clearance 
would result in lateral banging on a truck [sideframes]; that when 
inspecting car WSOR 501003 after the accident, he recognized that the truck 
bolster on the B-end of the car was the wrong size; and that the truck 
bolster had been installed at the NRCC repair shop. The shop manager also 
stated that excessive side bearing clearances would permit a car to 
experience greater harmonic action (rocking of the car side to side as it 
travels down a track), and that insufficient side bearing clearances would 
bind the side bearing rollers and therefore bind the truck (restrict the 
ability of a truck to swivel and therefore restrict the truck's ability to 
follow the contour of a track). The shop manager stated that before the 
accident, gib clearances were not routinely measured after repair work and 
that only a few car repairmen had gauges for measuring side bearing 
clearances. Since the accident and the implementation of a quality control 
program, he said that gib and side bearing clearances must be measured and 
each car repairman now had a proper measuring gauge. 

The shop manager said that new employees are trained by pairing them 
with experienced employees in a team concept, and that AAR manuals are used 
to identify wear tolerances and replacement requirements. When questions 
about replacement components cannot be confirmed by the AAR manuals, shop 
managers call the AAR or car component manufacturers for assistance. 

Other Car Repair Information . --Following the postaccident detection of 
the mismatched truck bolster on WSOR 501003, the owner of the Wisconsin & 
Southern Railroad told his car inspectors to look closely for mismatched 
bolsters on other cars. On May 10, 1989, Wisconsin & Southern Railroad car 
inspectors found a freight car with excessive gib clearances. This car, CNW 
69883, was loaded with sugar and offered at interchange by the Chicago and 
North Western Transportation Company. The sideframes at both the A- and 
B-ends had been damaged as a result of contact with the truck bolster 
(figure 17). 

The Safety Board sought historical car repair records to determine if 
the car had been previously derailed and if it was equipped with a correct 
bolster. No records of a previous derailment were found and no records to 
indicate a mismatched bolster were found. Wisconsin & Southern Railroad 
repair records indicate that identical type bolsters were replaced at the 
time of repair, May 25, 1989. 



43 




A-END 




B-END 

Figure 17. --Damage to the sideframes of car CNW 69883. 



44 

Freight Car and Train Inspection Information 

Freight Car Inspections . --The FRA requires predeparture inspections of 
freight cars in 49 CFR Part 215.13: "At each location where a freight car 
is placed in a train, the freight car shall be inspected before the train 
departs. This inspection may be made before or after the car is placed in 
the train." The regulation also states that if a designated inspector 
(designated by the railroad), defined in Section 215.11, is on duty for the 
purpose of inspecting at a location where the inspection is required, the 
designated inspector must determine whether the car is in compliance with the 
Federal safety regulations. Normally, a designated inspector is available at 
a railroad yard or interchange point. 

However, at a location where a designated inspector is not on duty (for 
example, at an industrial site where a traincrew picks up a freight car), 
the freight car must be inspected, at a minimum, for imminently hazardous 
conditions considered likely to cause an accident or casualty before the 
train arrives at its destination. Items that must be inspected if a 
designated inspector is not available are specifically defined in the 
regulations: 

(1) Car body lean: (a) leaning or listing to side, (b) sagging 
downward, (c) positioned improperly on truck, (d) object 
dragging below, (e) object extending from side, (f) door 
securely attached, (g) broken or missing safety appliance, 
(h) lading leaking from a placarded hazardous material car; 

(2) Insecure coupling; 

(3) Overheated wheel or journal; 

(4) Broken or extensively cracked wheel; 

(5) Brake that fails to release; and, 

(6) Any other apparent safety hazard likely to cause an accident 
or casualty before the train arrives at its destination. 

The FRA considers hazardous conditions associated with this inspection to be 
readily discoverable by a traincrew. 37 

The FRA requires designated inspectors to be qualified to inspect 
railroad freight cars for compliance and to have demonstrated to the railroad 
a knowledge and ability to inspect railroad freight cars. 38 The FRA has not 
provided a standard to measure demonstrated knowledge and ability for 
designated inspectors (usually carmen) or train crewmembers; the FRA allows 
carriers to set their own standards. The FRA expects the inspection to be 
sufficient to assure that freight cars, including side bearing clearances, 
are in compliance with Federal regulations before they are transported. The 



The brakeman on train D812-26 performed this inspection on the five 
cars added to the train in Warwick; he found no imminently hazardous 
condition. 



38 



Title 49 CFR 215.11, Designated Inspectors 



45 

FRA regulations do not address gib clearances between truck bolsters and 
sideframes. 

Designated inspectors also inspect freight cars for compliance with 
safety standards in the Field Manual of the AAR Interchange Rules adopted by 
the Association of American Railroads. Among other items, designated 
inspectors (carmen) look for broken, damaged, and noncomplying components on 
the freight car. Items to be inspected include side bearing clearances and 
broken or excessively worn truck components. Although carmen look for 
improper or mismatched parts, mechanical supervisors for CSX said that during 
a normal predeparture freight car inspection or interchange inspection, it is 
unlikely that a carman would identify excessive gib clearance resulting from 
a mismatched truck bolster and sideframes. Rather, CSX supervisors said 
carmen performing these inspections would be expected to identify excessive 
gib clearances resulting from damaged or excessively worn parts. 

Inspections of Car WSOR 501003 . --On October 30, 1988, the Chicago and 
North Western Transportation Company (CNW) first accepted WSOR 501003 from 
the NRCC repair shop after it was rebuilt. This was the first opportunity 
for a railroad to inspect the car and to identify excessive gib clearances or 
improper side bearing clearances before the car entered the transportation 
system. The car was accepted by the CNW, interchanged with the Wisconsin 
Central Railroad Company, and then interchanged with the Wisconsin & Southern 
Railroad Company. 

WSOR 501003 was then filled with sand and offered for transportation 
with its first load since the rebuild. The full car was picked up by the 
Wisconsin & Southern Railroad, interchanged with the Wisconsin Central 
Railroad, and then interchanged with the CSX Railroad, which delivered the 
full car in Akron. The car was emptied and returned over the same route, 
completing its first round trip between November 18 and December 14, 1988. 
The car was filled with sand and completed a second round trip over the same 
route between January 16 and February 5, 1989. WSOR 501003 was on the first 
half of its third round trip when the Akron derailment occurred. 

Car WSOR 501003 had been interchanged between railroads at least 
12 times, and 4 separate railroads had been involved with the movement of the 
car between the time it was rebuilt and the time of the Akron derailment. 
The car was subject to several FRA-required predeparture freight car 
inspections during this period; no exceptions were taken to excessive gib 
clearances. 

The last carman to inspect WSOR 501003 conducted a predeparture 
inspection of the car in Willard, Ohio, between 2:45 p.m. and 3:45 p.m. on 
February 26, 1989, before it left for Akron. The CSX carman indicated that 
he checked about half of the 60 cars in the train and found no cars to be in 
bad order. 39 He said that when conducting predeparture freight car 
inspections, he looks for defects involving safety appliances, wheels, 
springs (missing), wear and tear on castings, brake beams, air brake linkage, 



39 



A second carman checked the remaining cars, 



46 

air hoses, and air brake setup. He looks at side bearing clearance and 
roller bearing condition if he notices a car leaning, and also checks 
stenciling on the car to determine when a car was last lubricated and when 
the brakes were last serviced. In his 10 years' work experience, he had 
never issued a bad order on a car because of excessive gib clearance or 
defective side bearing clearance. He does not measure gib clearances during 
predeparture inspections and would not look for that type of problem unless 
his attention was directed to the matter, such as in a CSX notice. He 
carried a wheel gauge, a tape measure, wrench, and small pinbar during the 
car inspections but carried no other gauges. The carman also stated that he 
was issued a current copy of the AAR field manual, but that he does not carry 
it with him while inspecting cars; instead, he keeps it in a nearby shelter. 

After this accident, the CSX Detroit division managers completed 
refresher courses on freight car and train inspections, and they reviewed 
inspection requirements and procedures for all involved mechanical and 
transportation personnel on the division. Additionally, CSX mechanical 
department officers conducted unannounced audits at rail yards throughout the 
CSX system and found that at some locations side bearing inspections were 
inadequate and that gib clearances were checked only for wear and damage. 
As a result, mechanical department officers at CSX headquarters sent letters 
to division-level supervisors systemwide instructing them to reemphasize 
procedures for side bearing clearance inspections with carmen and to provide 
increased training to carmen at locations where inadequate side bearing 
inspections were found. Additionally, the mechanical department acknowledged 
that gib clearance inspections at rail yards must go beyond inspecting for 
wear and damage, and instructed its division supervisors to increase carmen's 
awareness of the need to check gib clearances as part of the overall truck 
integrity assessment program. CSX intends to accomplish improved inspections 
of gib clearances by reemphasizing to carmen the AAR Interchange Rules that 
set gib clearance requirements. 

Tests of Train Air Brakes . --The FRA requires train air brake tests at 
the initial and intermediate terminals to determine that the brakes are 
operating properly. 40 In accordance with the regulations, each train must be 
inspected and tested where a train is originally made up and after one or 
more cars are added to the train. The regulations require that brake pipe 
leakage must not exceed specified limits, and brakes must apply and release 
as required. When a carman or qualified person performs the test at the 
initial terminal, the person participating in the test or a qualified person 
who had knowledge that the test was made must notify the engineer that the 
initial terminal road train air brake test was satisfactorily performed. 
According to the FRA regulations, this notification must be made in writing 
if the traincrew reports for duty after the qualified person goes off duty. 

An initial terminal air brake test was performed on train D812-26 by 
carmen at the Willard yard, and that information was provided verbally to the 
traincrew. 



Title 49 CFR Parts 232.12 and 232.13. 



47 

The brakeman stated that after he inspected the five cars picked up in 
Warwick, air was applied to the five cars and he observed that the air brakes 
on all five cars released. The cars were then moved and set aside while 
other cars were dropped off. The brakeman said that afterward, the 
locomotives and four front cars were reattached to the five cars. The brake 
system was charged and he observed the brakes set and then release on the 
last car, indicating that they functioned properly. The brakeman estimated 
that it took 7 to 8 minutes to perform the air brake test. 

Track Information 

Track . --The derailment occurred 232 feet south of MP 16 on the 
northbound main track of the Akron Branch Line, which extended in a 
geographically north/south direction for 27.1 miles between Warwick and 
Hudson, Ohio. The Akron subdivision consisted of 11.6 miles of single main 
track, which was owned by Conrail and maintained by Conrail's Pittsburgh 
division, and 15.6 miles of double main track owned jointly by Conrail and 
CSX. Conrail owned the northbound track, on which the accident occurred, and 
CSX owned the southbound track. Under a joint facility agreement, all tracks 
were maintained by Conrail and jointly used by both railroads. Conrail 
operated one train per day over the tracks, and CSX operated about 20 trains 
per day over the tracks. At the time of the accident, CSX and Conrail were 
negotiating for the sale of the Akron Branch Line to CSX; however, the 
proposed sale had not been filed with the Interstate Commerce Commission. On 
February 5, 1990, the sale of the Conrail -owned portion of the Akron Branch 
Line was finalized and ownership was transferred to CSX. 

In accordance with the joint facility agreement, CSX paid Conrail for 
using the track on a car-mile basis. Those funds collected by Conrail were 
not dedicated to maintenance expenses on the Akron Branch Line; instead, the 
funds were placed in a corporate account and track maintenance expenses were 
allocated on a total system priority basis, a standard industry practice. 

In the direction of train movement, the northbound track was on the east 
side (right) of the southbound track. Running parallel to the northbound 
track, on the east side, was an auxiliary track that provided access to 
Conrail's South Akron yard. The South Akron yard consisted of 17 tracks 
located north (direction of train travel) of the accident site between MP 16 
and MP 15. The auxiliary track was about 10 inches lower (in elevation) than 
the northbound track. 

Starting about 1.2 miles south of the derailment site (MP 17.1), the 
track grade made a transition from a 0.56-percent ascending grade to a 
0.09-percent ascending grade, and then to a 0.73-descending grade in the last 
3,893 feet leading into the point of derailment. The derailment site was at 
the bottom of a vertical curve where the track made a transition to a 
0.65-percent ascending grade. 

In the same 1.2-mile area, the track curved 2 degrees 5 minutes to the 
left for 1,337 feet and then turned tangent for 1,034 feet. It then curved 
left through a 2 degree 16 minute compound curve for 1,815 feet, becoming 
tangent for 729 feet. The track then curved right through a 1 degree 



48 

30 minute curve for 1,349 feet to the point of initial derailment, and about 
185 feet beyond that point. Prescribed superelevation was 1/2 inch with 31- 
foot spirals on each end of the curve. 

Two turnouts were located at the extreme north end of the curve 
(figure 18). One was a trailing point turnout in a crossover connecting 
northbound track No. 1 to southbound track No. 2; the other was a facing 
point turnout connecting northbound track No. 1 to the auxiliary track on the 
east side of the main tracks. A Conrail assistant division engineer for 
track stated that although it is not a good operating practice to install 
turnouts in a Class 3 track curve (the track designation at this location), 
it is a common [industry] practice. 

The northbound track in the area of derailment was constructed of 
132-pound continuous welded rail that was laid new in 1978. The rails rested 
on double-shouldered tie plates and 8 1/2-foot wooden crossties. The rail 
was restrained by two rail-holding spikes and two plate-holding spikes per 
tie plate. In general, every other tie was box-anchored to control 
longitudinal rail movement, with some anchors intermittently missing. 
Anchors were applied to every tie for about 200 feet approaching and leaving 
the turnouts. Only 1/2 inch of longitudinal rail movement was measured at 
the derailment site. 

Track surface and alignment was maintained on crushed rock ballast. At 
several locations, mud had permeated the ballast section. Muddy track 
conditions of this type were evident in the two main track turnouts 
(figure 19A) and on a bridge about 403 feet south of the point of initial 
derailment (figure 19B). A slow order for 25 mph was in effect at MP 24.8 
on northbound track No. 1 (8 miles from the accident site) at the time of the 
accident as a result of defects in the rail surface profile caused by mud. 

Both turnouts on the northbound main track were constructed with 
132-pound No. 10 bolted rigid frogs and were 23 feet long. The trailing 
point turnout was equipped with standard, reinforced, 20-foot switch points, 
with the main track closure rail bolted in place with 36-inch joint bars. 
The facing point turnout, however, was equipped with 27-foot switch points, 
with the main track closure rail welded to the heel of the switch point and 
the toe of the frog. The bent stock rail was undercut to accommodate the 
Samson main track (straight) switch point. 

Description of Physical Evidence . --A postaccident inspection of the 
track approaching the point of derailment revealed discrepancies in the curve 
alignment and the track surface. The uniform alignment of the 1 degree 
30 minute curve was interrupted by segments of irregular alignment caused by 
the presence of two turnouts located in the body of the curve. Throughout 
the curve, the high rail was curve-worn. However, at the toe of the trailing 
point frog, the rail wear tapered onto the closure rail for 14 feet, where it 
gradually tapered to no curve wear. Twelve feet north of the point where the 
high rail wear ended, wear was visible on the low rail for about 17 feet. 



49 




CD 


ro 


3 


-o 




4-> 


•r- 


1 — 








"O 


00 


X 


c 


3 


z: 


o 


o 


re 


<_> 


s- 




CU 


S- 


„ s 


01 


<c 


ro 


IT) 


sz 






o 


^2 


ro 


c 




5- 


ra 


CD 


+-> 




c 








T3 




(_) 


C 


+J 


r0 


3 


c 


<+- 


O 


CD 




J3 


S 


CU 


x: 


CU 


+J 


-M 


> 




3 


o 


00 


O 

00 


E 


+J 




c 


c 






a> 


CO 


ro 


■a 




S- 


■i— 




+J 


o 






o 


^ 4- 


CO 


U 


o 



ti «- = 



il ° >- — 



<= 3 



00 D)+ J Q. 

•r- S-M- 

s- o "+- 



50 





Figure 19. --Muddy track conditions on the northbound track: (A) at 
the turnouts, and (B) at the bridge south of the derailment site. 



51 

Segments of irregular alignment were also present in the facing point 
switch in the area of the switch, the closure rail, the frog, and behind the 
frog. 

Track surface through both turnouts was about 1/2 inch to 1 1/4 inch 
above the designated elevation; these differences were in compliance with 
Federal track standards. 

The first indication of a derailment was a flange mark on the base of 
the rail 47 feet behind the point-of-switch, in the facing point turnout 
(figure 20). This mark was on the right (field side) of the low (east) rail 
and was accompanied by a similar mark on the right (gauge side) of the high 
(west) rail. In the direction of train movement, the wheels derailed to the 
right side. There were only two marks representing one derailed axle. The 
marks progressed through the turnout and astraddle the frog; there were no 
marks on the guard rail. Marks indicated that a derailed wheel on the right 
(turnout) side struck the joint bars at the heel of the frog and climbed over 
the turnout rail. Immediately north of the frog, two distinct marks were 
visible in the center of the track between rails: one about 13 inches from 
the high rail and about 2 inches wide; the other about 25 inches from the 
high rail and about 3 inches wide. Only one mark was visible on the east 
side of the low rail near the end of the ties. The marks progressed 40 feet 
north of the frog to the point where total track damage occurred. 

The facing point switch was found in good adjustment and properly fitted 

to the undercut stock rail. Rail braces were tight and track gauge at the 

point of switch was 56 7/8 inches. 41 The switch rods were not bent, nor 
were there any marks on the Samson switch point. 

An abrasion was found on the receiving end of the guard rail in the 
trailing point turnout. The abrasion was on the guard face wing, about 1 1/2 
inches beyond the area of normal contact between the back side of the car 
wheels and the guard rail. The abrasion was oxidized with a brightly 
colored rust, consistent with recent origin. The source of the abrasion 
could not be determined. Additionally, rail corrugation, caused by wheel 
forces, was found near the bridge immediately south of the accident site (see 
figure 19B). 



According to 49 CFR 213.53, track gauge must be maintained between 
56 inches and 57 3/4 inches. 



52 




Figure 20. --Flange mark on base of rail 



Track Inspections .--The Akron subdivision was designated by Conrail as a 
Class 3 track and was therefore required to meet Federal track safety 
standards for that class. 42 These standards required the carrier to inspect 
the track twice weekly, with at least one calendar day interval between 
inspections. In the 30-day period prior to the derailment, the Conrail track 
inspector inspected the subdivision in its entirety eight times. During 
these inspections, nine FRA defects (items that were not in compliance with 
FRA track standards) were found on northbound track No. 1: one broken joint 
bar, five joints with sheared bolts, and three joints with inadequate bolts. 
Other track defects were noted that were not in violation of Federal track 
safety standards. 

A track geometry car last evaluated the territory on December 13, 1988. 
Three FRA defects were detected within 1 mile south of the derailment site: 
irregular crosslevel at MP 16.2, MP 16.6, and MP 16.8. 



42 According to 4 9 CFR 213.9, the maximum allowable operating speed for 
a freight train is 4 mph for a Class 3 track, 2 5 mph for a Class 2 track, 
and 10 mph for a Class 1 track. If a segment of track does not meet all of 
the requirements for its intended class, it is reclassified to the next 
lowest class of track for which it does meet all of the requirements. In 
such an instance, a railroad may issue a slow order until conditions are 
corrected. 



53 

The Akron subdivision was last inspected by an FRA track inspector for 
compliance with Federal track safety standards on January 30, 1989. Several 
inches of snow covered the ground at the time of the inspection. During the 
inspection, one rail brace defect and nine track bolt defects were detected 
on northbound track No. 1. Two of the bolt defects were in the immediate 
area of the derailment and records indicate that repairs were made prior to 
the accident. 

After the accident, between March 1 and 3, 1989, State and FRA track 
inspectors identified 75 track defects on northbound track No. 1 between MP 
11.8 and MP 27.2: 30 track bolt defects, 29 locations of fouled ballast, 43 4 
tie defects, 4 track gauge defects, 5 defective joint bars, 2 irregular 
crosslevel defects, and 1 switch point stop defect. Some of the defects near 
the accident site included the following: 

Fouled ballast from MP 16.4 to 16.7 causing 
track profile to deteriorate; 

Crossties not effectively distributed to 
support a 39-foot segment of train over 
bridge 16; 

Fouled ballast at north end of bridge 16 
causing track profile to deteriorate; 

Fouled ballast at MP 15.8 causing track profile 
to deteriorate (about 100 feet); and 

Fouled ballast at MP 15.2. 

According to the regional track engineer of FRA's Region 2, in which the 
Akron subdivision is located, the track was not in compliance with Class 3 
requirements, based on the finding of 137 defects on the northbound and 
southbound tracks between MP 11.6 and MP 27.2. He stated that because of 32 
joint-bolt defects in which there were fewer than two bolts in each end of a 
joint, the condition of the track was good only for Class 1. Other items, 
such as missing or loose frog bolts and fouled ballast, would permit the 
tracks to be classified as Class 2. 

In Ohio, there are two FRA track inspectors and three certified State 
inspectors who work for the Public Utilities Commission of Ohio. Each 
inspector is responsible for a distinct territory. The FRA inspector in 
Cleveland, Ohio, is responsible for inspecting the Akron Branch Line. 



Ballast, the material beneath the ties, may become fouled 
(contaminated) by different materials, usually dirt or mud percolating up 
from the subgrade. Fouled ballast, in and of itself, may not constitute an 
FRA defect unless the condition affects track geometry or tie support 
conditions. 



54 

Track Maintenance . --A three-person Conrail maintenance crew, consisting 
of one foreman, one truck driver, and one trackman/machine operator, assisted 
by a welder and helper, was assigned to maintain the Akron subdivision 
(27 miles). The crew was also responsible for maintaining the frequently 
used Cleveland Line (31 miles), and the infrequently used Rittman Secondary 
(11 miles) and Freedom Secondary (33 miles). The efforts of the crew were 
supplemented by mechanized, extra crew activity, which was allocated 
annually. 

In 1987, $512,000 was allocated and spent on the Akron subdivision 
northbound track rehabilitation program: 8,136 ties were installed and track 
was surfaced between MP 18.4 and MP 27.2. In 1988, Conrail track maintenance 
personnel requested $325,000 for tie renewals and 9 miles of track surfacing 
between MP 11.6 and MP 18.4 on the northbound track; all funding for this 
rehabilitation program was denied by Conrail management. The only extra crew 
activity in 1988 occurred in December: a track surfacing crew spot-tamped 
the track at selected locations. Track maintenance personnel requested for 
1989 $830,000 in the rehabilitation program for track surfacing and tie 
renewals on northbound track No. 1 between MP 11.7 and MP 27.1. According to 
Conrail 's assistant engineer of track, funding is requested each July for the 
following calendar year; the requested 1989 funding had not been approved at 
the time of the accident or by July 1989 at the time of the Safety Board's 
public hearing. 

Training and Emergency Preparedness 

Training of Railroad Personnel .--The traincrew had not been provided 
emergency response training outside of routine CSX operating rules classes. 
The classes instruct employees on their responsibilities in the event of an 
accident involving hazardous materials. This training includes a 45-minute 
video tape on the proper handling of hazardous materials. 

Operating Rule 74 requires the conductor or other CSX personnel at an 
accident site involving hazardous materials to initiate actions necessary to 
ensure the public safety, and to protect property and the company's 
interest. In part, the rule requires employees to take the following actions 
as soon as possible and if it is safe to do so: 

(a) Rescue the injured, remove them to a safe area, administer 
first aid, and call for assistance; 

(b) Survey the scene and adjacent area, assess the situation, 
ascertain the conditions, including identifying cars/trailers 
containing hazardous materials involved in the emergency or in 
the immediate area, and notify the proper authority by the 
quickest means available; 

(c) Protect life and property. This may require evacuation of 
people from the area; fire fighting; removal of cars or 
containers and their contents; and 



55 

(d) In the event that the emergency involves spillage, loss of 
hazardous materials or fire, the conductor or his designee 
must notify or request the chief dispatcher to notify the 
nearest EMERGENCY RESPONSE GROUP, such as the fire and police 
departments and medical rescue, among others. The conductor 
or his designee must remain at the scene until the arrival of 
the Emergency Response Group or until released by the proper 
authority. 

Operating Rule 74-A provides evacuation guidelines to the traincrew, and 
Rule 74-C requires the conductor to take the initiative to identify himself 
to the emergency response persons when they arrive at the scene. The rule 
also requires the conductor to furnish emergency response persons with 
information obtained from the waybills and the train consist concerning 
hazardous materials in the train and to provide his knowledge of the current 
conditions. Operating Rule 74 cautions employees that information 
identifying the contents of cars involved in the emergency and exact 
conditions (whether fire, explosion, fuming or leaking of hazardous 
materials) is vital in determining the required corrective action. A current 
copy of these rules must be carried by traincrews while on duty. 

The CSX had prepared efficiency tests to determine if traincrew actions 
were performed in accordance with company operating rules; the efficiency 
tests include criteria for evaluating train and engine crew performance when 
accidents occur that involve hazardous materials. The efficiency tests 
inquire if the conductor identified himself to emergency response persons, 
furnished waybill and consist information, and provided knowledge of current 
conditions. 

A CSX trainmaster who responded to this accident told Safety Board 
investigators that he had never performed efficiency tests on train 
crewmembers regarding their responsibilities and actions taken as a result of 
a hazardous materials emergency. (The trainmaster said that he believed that 
this could be done only after an emergency.) Although he said he had not 
performed a formal efficiency test on the traincrew after this accident, he 
took no exception to their actions relative to the rules because he believed 
that the conductor, although separated from the fire chief because of the 
fire, provided necessary information regarding hazardous materials involved 
in the derailment to the brakeman who was with the fire chief. 

According to a CSX corporate official, since the accident, all CSX 
Detroit division managers have been provided enhanced training on response to 
rail transportation emergencies involving hazardous materials. Further, 
division supervisors are now providing additional instruction and testing to 
traincrews on the matter. CSX is also considering the production of 5- to 10- 
minute, specific topic video tapes on hazardous materials rules and 
emergency response for use in operating rules classes. 

Emergency Preparedness . --The Akron Fire Department implemented its 
hazardous materials operations plan, which served as an onscene checklist 
for each fire department unit. The plan included guidelines for other 
emergency response agencies involved in the incident. 



56 

The City of Akron and Summit County, in which Akron is located, had an 
emergency operations plan, adopted in July 1988, and administered by the 
Summit County Office of Emergency Management. The plan was used to help 
facilitate the evacuation and provide guidance to first responders and 
supporting agencies. Additionally, the Akron Police Department had a general 
order in effect, which described the actions to be taken by its members, 
relevant to hazardous materials incidents. 

The fire department conducted hazardous materials incident drills among 
its firefighters on an instantaneous order from the hazardous materials 
chief; the drills had not included participation by the railroads operating 
within the city. The CSX division manager for hazardous materials, however, 
had conducted in November 1988, an 18-hour class for the fire department that 
included hands-on instructions for hazardous materials incidents involved 
with railroad tank cars. In October 1984, Conrail had provided 16 hours of 
classroom and hands-on hazardous materials training for the fire department. 

In October 1988, the B.F. Goodrich Chemical Company and Conrail held a 
simulated leak of volatile chemicals inside the plant facility while a 
Conrail train and crew were onsite. In 1987, the B.F. Goodrich Chemical 
Company provided tours of their facilities to Conrail 's South Akron yard 
personnel. Additionally, the B.F. Goodrich Chemical Company had provided the 
fire department with information about hazardous materials used in the 
facility and had cooperated with the fire department in preparing emergency 
plans. 

In November 1989, after this accident, a cooperative venture between 
CSX, the fire departments and rescue agencies of Summit County, and the City 
of Akron resulted in 6 days of comprehensive orientation, training, and 
field exercises on the handling of hazardous materials incidents in rail 
transportation including tank car, tank truck, and locomotive emergencies. 
In addition, a major chemical manufacturer and a hazardous materials 
emergency response company made presentations to fire department personnel, 
and CSX officials met with city leaders to assure that a cooperative 
relationship was fully in place in Akron. The CSX believed the training 
effort to be successful and plans to conduct similar training programs at 
other locations on its system. 

CSX is currently developing a computerized hazardous materials data 
program that will be able to provide pertinent historical information about 
hazardous materials movements to local emergency planning committees and 
agencies in communities through which CSX travels. Additionally, the CSX and 
the railroad industry are researching methods of providing aboard each train 
real-time documentation of train consists. 

Meteorological Information 

On February 23, 1989, conditions at Akron included light snow showers 
and temperatures between 10 °F and 19 °F. On February 24, the maximum 
temperature was 23 °F and the minimum was 5 °F; there was no precipitation. 
On February 25, the maximum temperature was 38 °F and minimum was 8 °F; there 



57 

was no precipitation, and skies in the Akron area changed from clear to 
overcast throughout the day. 

On February 26, the skies over Akron were overcast throughout the day 
with light snow. Total snow accumulation was 2.3 inches (water equivalent of 
0.2 inch). The maximum temperature was 34 °F and the minimum was 26 °F. 

During the following 4 days, temperatures varied between 14 °F and 36 °F 
and only a trace of snow was recorded. On March 3, the temperature increased 
to 52 °F and overcast skies changed to partly cloudy. 

Toxicological Information 

All four crewmembers were tested for drugs and alcohol at the Akron City 
Hospital between 12:04 a.m. and 12:31 a.m., February 27, 1989. Blood and 
urine samples were obtained and sent to the Center for Human Toxicology, 
University of Utah, Salt Lake City, Utah. All samples tested negative for 
drugs and alcohol . 

Tests and Research 

Event Recorders . --Event recorder data packs were removed from CSX units 
6124, 6675, 6850, and 6038--the locomotives for train D812-26--and were sent 
to the Safety Board in Washington, D.C., for readout and evaluation. The 
data packs came from three-event recorders, which record elapsed time and 
wheel revolutions for generating unit speed and distance traveled over 
elapsed time. 

Strip charts were generated to show train movement, time, distance, and 
speed. The following information was produced: 

The maximum speed attained by unit 6124 just prior to rapid 
deceleration was 43.8 mph (with an error tolerance of +0.1 mph); 

The train traveled about 1,195 feet during deceleration; and 

After the train stopped following the rapid deceleration, unit 
6124 remained stopped for about 8 minutes 35 seconds, and then 
traveled a distance ranging between 394.6 and 427.4 feet during a 
period of about 56 seconds. 

The Rail Safety Improvement Act of 1988 mandates rules requiring event 
recorders. The FRA does not require the use of event recorders. 

Tests of Locomotives . --CSX locomotives 6124, 6038, 6850, and 6675 were 
inspected and tested on March 4, 1989, by CSX, FRA, and Safety Board 
personnel at the CSX mechanical shops in Cumberland, Maryland. Each unit was 
tested or checked to (1) calibrate the speed recorder, and (2) calibrate the 
speed indicator. All speed recorders and speed indicators were within the 
FRA requirement of +3 mph for speeds of 30 mph or less and +5 mph for speeds 
above 30 mph. At 40 mph, the speed indicator on unit 6124 displayed 39 mph 
while the speed recorder recorded 40 mph. 



58 

The power brake interlock was checked on the four locomotives from the 
lead locomotive 6124. Unit 6124 was a GP40-2 built in 1975. The remaining 
three locomotives were older GP40 units. Unit 6124 had a unique trainline 
emergency feature designed to give an engineer 18 seconds in which to release 
(bail off) his independent or locomotive brakes and still apply power in an 
effort to pull or break away from the rest of the train and to prevent any 
cars from piling into the locomotives and crew. After 18 seconds, the 
engine drops to idle and the power cannot be reinitiated by throttle 
modulation until the power control relay is reset, which requires a 90-second 
delay after the trainline air is recovered to the original feed valve setting 
on the control locomotive. 

The test verified that the power control relay functioned as designed 
and did not actuate the power brake interlock for 18 seconds after an 
emergency brake application occurred, such as a trainline separation or 
excessive leakage. 

Land-Use Information 

The City of Akron does not restrict the use of land adjacent to mainline 
railroad tracks through minimum setback requirements (buffer zones) or by 
prohibiting the permanent or temporary storage of hazardous materials 
adjacent to mainline tracks. 

The CSX, however, does impose some restrictions on the use of land it 
controls adjacent to mainline tracks. On land it owns and leases to others, 
the CSX prohibits the storage of liquefied petroleum gas in permanent tanks, 
with a capacity exceeding 12,000 gallons, within 200 feet of mainline tracks 
unless special permission is granted. Additionally, the CSX restricts the 
permanent storage of flammable and combustible liquids near its mainline 
tracks. For example, storage tanks with a capacity between 18,000 and 50,000 
gallons must be set back 75 feet from mainline tracks, and storage tanks 
exceeding 250,000 gallons capacity must be set back 200 feet. 

An Interindustry Rail Safety Task Force established by the Chemical 
Manufacturers Association (CMA) and the AAR, and consisting of senior 
managers representing the chemical, and railroad and tank car manufacturing 
industries recently made several recommendations to the chemical and railroad 
industries, which have been adopted by the CMA and the AAR. In conjunction 
with those recommendations and on January 4, 1990, the AAR issued Circular 
No. 0T-55, "Recommended Railroad Operating Practices for the Transportation 
of Hazardous Materials." In part, Circular No. 0T-55 establishes proposed 
separation distances for loaded tank cars and storage tanks from mainline 
Class 2 track or better. The recommended separation distances, which apply 
to storage on railroad property and on chemical company property located 
close to railroad mainline tracks, are as follows: 



59 

Combustible Liquid, Poison Inhalation Hazard 

Corrosive Material (Packing Group I) and 

Activity and ORMs 44 All Other Hazard Classes 

(feet) (feet) 

Loading or unloading: 

(a) If conditions permit 50 100 

(b) Minimum distance 25 50 

Storage of loaded tank cars 25 50 

Storage in tanks (permanent) 

(a) If conditions permit 50 100 

(b) Minimum distance 25 50 

The recommendation on separation distances states that "with regard to 
existing facilities maximum reasonable effort should be made to conform to 
this standard taking into consideration cost, physical and legal 
constraints." 

The recommended separation distances were made, in part, after members 
of the task force made graphs from Safety Board accident data on the maximum 
distance equipment traveled from track center in accidents. The group's 
analysis of the graphs showed that in about 50 percent of the railroad 
accidents in the Safety Board data base, equipment traveled more than 40 feet 
from the track; in about 2 percent of the accidents, equipment traveled more 
than 100 feet from track center. However, the task force was advised by its 
members to interpret the results with caution because (1) the accidents 
selected for analysis were not a random, independent sample of all train 
accidents but were believed to be more serious than average, and (2) the 
analysis does not control for steep embankments adjacent to track that might 
cause equipment to travel farther. 

Other information considered by the task force included (1) a series of 
recommendations made by the Safety Board to the AAR on April 28, 1982, that 
addressed the vulnerability of hazardous materials stored near tracks, 45 and 
(2) consistency with nationally recognized safety standards. Standards 
considered by the task force included the American National Standards 
Institute K-61. 1-1989 for anhydrous ammonia; National Fire Protection 
Association (NFPA) No. 30, "Flammable and Combustible Liquids Code"; and NFPA 
Standard No. 58, "Storage and Handling Liquefied Petroleum Gases." 



DOT hazard classification for other regulated materials (ORM). 

45 Safety Recommendations 1-82-1 through -7 issued April 28, 1982, are 
addressed in the analysis of this report. 



60 

ANALYSIS 

General 

The four crewmembers of train D812-26 were qualified for their positions 
by CSX. Postaccident tests of the airbrakes indicate that they were 
functioning properly and were, therefore, not a factor in this accident. 
Drugs and alcohol were not factors in this accident. Weather also was not a 
factor in the accident. 

The Derailment 

The physical evidence indicates that the point of derailment occurred 
near the switch in the facing point turnout on the northbound mainline track. 
Postaccident inspection of the cars involved in the derailment revealed that 
only one car exhibited the heavy rail burn marks (which were the result of 
the derailment) seen on the left sideframe of the B-end truck of car WSOR 
501003. Because this physical evidence strongly suggested that WSOR 501003, 
the 13th car in the train behind the locomotives, was the first car to 
derail, the investigation examined various factors that could have led to the 
derailment of this car including the mechanical condition of WSOR 501003, the 
operation of the train as it approached the derailment site, and the track 
conditions in the derailment area. 

The postaccident inspection of WSOR 501003 revealed (1) that the truck 
bolster on the A-end of the car was different from the truck bolster on the 
B-end, (2) that the gib clearance between the bolster and truck sideframe on 
the B-end truck was more than 3 inches, exceeding the limits listed in the 
AAR Interchange Rules, and (3) that the side bearing clearances exceeded the 
limits listed in the FRA regulations and the AAR Interchange Rules. Because 
the original equipment was used in reconstructing the cars after the 
derailment and because there were no broken parts associated with the truck 
bolsters and sideframes on WSOR 501003, there should have been no variation 
in the measurements of side bearing clearances and gib clearances taken 
before or after the accident. Therefore, the Safety Board concludes that the 
excessive gib clearance on the B-end truck and the out-of-limits side bearing 
clearances on WSOR 501003 existed before the derailment. There was no 
evidence of preaccident mechanical deficiencies with the other cars involved 
in the derailment. 

Limits on gib clearances are set to provide the optimum maneuverability 
of the truck to follow the contour of the rail as the wheels of a car follow 
the track. Insufficient gib clearance will tend to cause the truck to bind 
as the rail car enters a curve, whereas excessive gib clearance results in 
the truck not being guided adequately (too loosely) through the curve. Out- 
of-limits side bearing clearances can result in the car rocking excessively 
from side to side, particularly if the car traverses any irregular track 
conditions. 



61 

Postaccident inspection of the track in the derailment area revealed 
segments of the track in which the higher rail was worn. This is evidence of 
irregular alignment, probably as a result of the two turnouts located in the 
curved area of the track. Areas of fouled ballast approaching the point of 
derailment also existed. 

There was no evidence to indicate that the train was being operated any 
differently than it had been on previous occasions, and, based on the 
engineer's testimony that the throttle was in the idle position, there would 
not have been any operationally induced in-train dynamics, such as slack run- 
in as a result of brake applications. The three-event recorder indicated 
that the train was probably traveling about 43 mph at the time of 
derailment, or 3 mph over the maximum allowable speed. As a result of fouled 
ballast in the area approaching the derailment site, WSOR 501003 may have 
experienced some rocking motion that would be compounded or exaggerated by 
the excessive side bearing clearances. The lower the speed of the train, the 
more time there is available for a car, after traversing an area of fouled 
ballast, to correct itself or "settle down" before reaching another area of 
track irregularity. Because the area of fouled ballast near the bridge was 
more than 400 feet from the point of derailment, there was sufficient time 
for WSOR 501003 traveling at 43 mph to "settle down" before reaching the 
turnouts in the curve. However, just before the turnouts, the ballast was 
again fouled and this may have permitted the car to again begin rocking as 
it entered the first turnout. 

Irregularities in the track surface and alignment affect normal lateral 
and vertical forces of wheels moving along the rails; on curves, the effects 
of these irregularities become more pronounced. As car WSOR 501003 entered 
the curve, vertical wheel loading shifted as a result of a transfer of weight 
to the wheels on the outside of the curve and, in combination with the 
irregular track alignment (turnouts in the curve), caused lateral movement of 
the car (or slapping of the wheels against the rails). Because of the 
excessive gib clearance on the B-end truck (the lead truck in the car on this 
trip), the truck was being loosely guided through the curve and the turnout; 
this action in combination with the lateral and vertical movement of the car, 
exacerbated by the out-of-1 imits side bearing clearances, probably caused the 
lead wheels to derail. The Safety Board concludes, therefore, that the 
excessive gib clearance on the B-end truck and the out-of-limits side bearing 
clearances on car WSOR 501003 combined with the fouled ballast and the 
irregular alignment of the track in the turnouts caused the derailment of 
train D812-26. 

The Safety Board attempted to determine why WSOR 501003 had not derailed 
during the previous two trips over the same territory. However, the Safety 
Board could not determine which end of the car was the lead end during the 
previous trips, the position of the car in the trains during the previous 
trips, the speed at which the trains were being operated during the previous 
trips, or the condition of the track during the previous trips--all factors 
that could have affected the dynamics of the car. The Safety Board, 
therefore, could not definitively determine why the car did not derail on the 
previous trips. However, it is likely that all the conditions that existed 
on February 26, 1989, did not exist during either of the previous trips. 



62 
Event Recorders 

The event recorders installed on the four CSX locomotive units recorded 
only elapsed time, locomotive speed, and distance traveled; however, multi- 
event recorders are readily available that record other parameters including 
traction motor current, throttle position, locomotive braking, and direction 
of travel. Had the units been equipped with multi-event recorders, Safety 
Board investigators would have been able to obtain more information 
regarding the manner in which the train was being operated approaching the 
accident site, particularly throttle positions and the timing of brake pipe 
pressure reductions. Further, without the multi-event recorders, 
investigators were unable to corroborate crew testimony about certain aspects 
of how the train was being operated immediately before the derailment. 

The Safety Board has documented its position regarding the mandatory use 
of event recorders in the railroad industry in previous accident 
investigations 46 and through the issuance of safety recommendations to the 
industry and the Federal Railroad Administration. Further, the Safety Board 
has stated in previous reports of accident investigations that the Rail 
Safety Improvement Act of 1988 mandates rules requiring event recorders and 
does not give the FRA freedom to decide whether Federal regulatory 
intervention on this subject is necessary. Consequently, the Safety Board 
has recommended that the FRA "expedite the rulemaking requiring the use of 
event recorders in the railroad industry" (Safety Recommendation R-89-50). 
The FRA's position, as stated on March 30, 1990, in its most recent response 
to Safety Recommendation R-89-50, is that the FRA is "analyzing. . .information 
to determine if a Federal requirement for event recorders is cost beneficial, 
and if so, how a rule is to be implemented." While the lack of effective 
action by the FRA continues to cause the Safety Board concern, it should be 
noted that the FRA administrator has agreed that some type of recording 
device should be required on trains, and a proposed rule is currently being 
developed by the FRA. In view of FRA's effort to proceed with rulemaking 
activity, Safety Recommendation R-89-50 will be classified as "Open- 
Acceptable Response." 

Track Inspections 

About 1 month before the accident, the Akron subdivision was inspected 
by an FRA track inspector to determine compliance with the Federal track 
safety standards. During this inspection, one rail brace defect and nine 
track bolt defects were noted on the northbound track. Two days after the 
accident, the same area was inspected by State and Federal track inspectors, 



46 (a) National Transportation Safety Board. 1989. Head-on collision 
between Iowa Interstate Railroad Extra 470 West and Extra 406 East with 
release of hazardous materials, near Altoona, Iowa, on July 30, 1988. 
Railroad Accident Report NTSB/RAR-89/04. Washington, DC. 98 p. (b) 
National Transportation Safety Board. 1990. Derailment of Southern Pacific 
Transportation Company freight train on May 12, 1989, and subsequent rupture 
of Calnev petroleum pipeline on May 25, 1989, San Bernardino, California. 
Railroad Accident Report N T SB/ RAR - 90/02 . Washington, DC. 193 p. 



63 

and 75 track defects were noted including 30 track bolt defects and 
29 locations of fouled ballast. Certain track defects, such as missing bolts 
and defective joint bars, are readily determinable to track inspectors, 
whereas other track conditions, such as fouled ballast, require a more 
subjective evaluation. The discrepancy in the number of defects noted 
1 month before the accident and immediately after the accident is of concern 
to the Safety Board. Although the track in the immediate area of the 
derailment may have been in compliance with Class 3 standards, the track in 
the general area north and south of the accident site was determined by an 
FRA inspector after the accident to be out of compliance with the Class 3 
standards. Although the Safety Board acknowledges that track conditions may 
deteriorate during a 1 -month period, the Safety Board does not believe that 
conditions deteriorated rapidly enough in that short period to account for 
the discrepancy in the defects noted after the accident. Several inches of 
snow covered the ground when the first inspection was conducted, which raises 
concerns about the adequacy of conducting an inspection under those 
conditions. The snow cover may explain why many defective track conditions 
went unnoticed. However, because there is a certain amount of subjectivity 
in evaluating some track conditions, it is also possible that defective track 
structure conditions are not being uniformly noted in the Akron subdivision. 
Whatever the reason for the discrepancy in the track defects noted, the 
Safety Board is concerned about the quality and thoroughness of the FRA's 
inspection of the track prior to the accident. Accordingly, the Safety Board 
believes that the FRA should evaluate the adequacy of track inspections being 
conducted on the Akron subdivision and institute necessary changes to ensure 
proper inspections. 

Track Maintenance 

The results of the investigation suggest that the maintenance program in 
place for the tracks in the accident area had reached a point of being 
reactive rather than preventive, possibly because of the pending sale of the 
northbound track. All funding requested by Conrail track maintenance 
personnel for 1988 track rehabilitation had been denied. The only extra crew 
activity occurred in December 1988 when a track surfacing crew spot-tamped 
the track at selected locations. At the time of the accident, the funding 
request by maintenance personnel for 1989 track rehabilitation had not yet 
been approved. The condition of the track in the area after the accident, as 
documented by track inspectors, and the fact that no funding had been 
approved for track rehabilitation for more than a year, suggests that the 
pending sale of the northbound track to CSX may have affected Conrail 's 
decision to not allocate the necessary resources for the rehabilitation. 
Conrail 's decision to delay rehabilitation of the track or not to place a 
slow order on the track in the accident area contributed to the cause of the 
accident. 

Freight Car Inspections 

Car WSOR 501003 had been inspected on several occasions during 
predeparture inspections, as required by Federal regulations, and when the 
car had been interchanged between railroads, in accordance with the AAR 
Interchange Rules. WSOR 501003 had been inspected both at locations where a 



64 

designated inspector was and was not on duty. Given that only six items 
must be inspected for determining imminently hazardous conditions if a 
designated inspector is not on duty, the Safety Board believes that a train 
crewmember could not have been expected to detect the excessive gib and out- 
of-limits side bearing clearances on WSOR 501003. 

However, the Safety Board believes that the excessive gib and out-of- 
limits side bearing clearances should have been detected by a designated car 
inspector in the car inspection system. The Safety Board is concerned that 
the excessive gib and out-of-limits side bearing clearances were not detected 
during the 12 or more interchange inspections and several predeparture 
inspections. Because detection of the excessive gib and out-of-limits side 
bearing clearances may have prevented the accident, the failure of car 
inspectors to do so is considered causal to the accident. The failure to 
detect the excessive gib clearance may have been in part because of the 
tendency of car inspectors to look for damaged or excessively worn parts, 
such as those later found on CNW 69883, rather than mismatched bolsters or 
sideframes. The testimony of the car inspector who last inspected WSOR 
501003 prior to the accident also suggests, however, that car inspectors are 
not trained or instructed to look for excessive gib clearance. Further, 
testimony by the CSX supervisory personnel indicates that the car inspector 
would not have been expected to notice the excessive gib clearance. 

The Safety Board notes the corrective action taken by CSX following the 
accident to improve inspections of side bearing and gib clearances. The 
Safety Board believes that the additional instructions and training regarding 
gib and side bearing clearance inspections should help to detect the type of 
excessive clearances that were evident in this accident. The Safety Board 
also believes that the Association of American Railroads should inform its 
members of the circumstances of the accident, emphasizing the need for car 
inspectors to check side bearing and gib clearances during inspections. 

Northern Rail Car Corporation Shop Procedures 

The many deficiencies noted during the postaccident inspection of the 
rebuilt WSOR series cars—including out-of-limits side bearing clearance, no 
side bearing clearance, loose side bearing top plates, a cracked side 
bearing top plate, mismatched truck sideframes, and excessive gib clearance 
--indicate that quality control procedures at the Northern Rail Car 
Corporation (NRCC) facility at the time the 36 cars were being rebuilt were 
inadequate. The Safety Board believes that procedures should have been in 
place at the NRCC to detect the deficiencies in the WSOR series cars. 
Therefore, the lack of adequate rebuild and quality control procedures at 
the NRCC is considered causal to the accident. The Safety Board notes, 
however, that following the accident, NRCC implemented new procedures 
including (1) the use of a quality control inspection sheet in conjunction 
with the repair of all cars to note, among other items, the measurement of 
gib and side bearing clearances, and (2) more frequent inspections by two 
supervisors as cars are being repaired. These improvements in shop 
procedures should help to eliminate the type of deficient repairs that were 
made during the rebuild project. 



65 

Association of American Railroads' Inspection of Freight Car Repair Shops 

In February 1987, the AAR notified the NRCC freight car repair shop of 
deficiencies noted the month before during an inspection of the facility by 
an MID inspector; the AAR advised the shop to correct the deficiencies and 
requested to be informed when such action was taken. The AAR, however, did 
not followup to determine if the deficiencies had been corrected. The Safety 
Board believes that for the inspections to be effective, followup action is 
necessary to determine if deficiencies noted have been corrected. 

Association of American Railroads' Inspection of Rebuilt Cars 

AAR's procedures require that one or more rebuilt cars in a project be 
inspected by an MID inspector to determine if the car(s) meets all current 
safety standards and Interchange Rules. The Safety Board received 
conflicting testimony as to how many cars were available for inspection when 
the MID inspector visited the NRCC facility to inspect one of the cars from 
the project. The MID inspector stated that he inspected only one car because 
it was the only one available. The owner of the facility testified, however, 
that he believed five cars were available for inspection because of the 
numbering sequence of the cars. Regardless of the number of cars available 
at that time, only one car was inspected, and the MID inspector determined 
that it met all appropriate standards and rules and that all remaining cars 
should be rebuilt to the same specifications. The investigation revealed, 
however, that all remaining cars were not rebuilt to the same specifications. 
The Safety Board believes that one visit to the facility and the inspection 
of only one car is not sufficient to determine if all cars being rebuilt in 
the project meet the appropriate standards and rules. The AAR, therefore, 
should develop and implement procedures to provide for adequate followup 
inspections of cars during a rebuild project. 

One of the cars in the rebuild project, WSOR 501017, as later noted by 
another MID inspector in Chicago, did not conform to AAR rules because it was 
equipped with threaded fittings in the air brake line and did not comply with 
FRA safety regulations because it was equipped with side ladders extending 
from the bottom to the top of the car. All 36 cars in the rebuild project 
were similarly equipped, and the AAR required the Wisconsin & Southern 
Railroad to correct these deficiencies. The Safety Board is concerned that 
these two areas of obvious noncompliance were not noted by the MID inspector 
who inspected WSOR 501032 at the NRCC facility and believes that the 
thoroughness of initial car inspections should be addressed in AAR's 
inspection procedures for rebuild projects. 

Tank Car Performance 

Two of the tank cars involved in the accident received hard hits in the 
top half of the tank head, yet no breaching of those tanks occurred. Several 
tank cars were fully engulfed in fire, yet there was no violent rupture from 
exposure to heat. All of these tank cars were equipped with head shields and 
thermal protection. Additionally, all nine tanks were equipped with shelf 
couplers, some of which broke during the derailment and resulted in strikes 
to tank car heads. The Safety Board believes that this accident 



66 

demonstrates the safety benefits of full tank car head shield and thermal 
protection even when the cars are equipped with vertical restraint (shelf) 
couplers. 

On May 15, 1990, the Research and Special Programs Administration issued 
an Advance Notice of Proposed Rulemaking (ANPRM) "Specifications for Tank Car 
Tanks," [RSPA Docket No. HM-175A; Notice No. 90-8]. One of the issues 
addressed in the ANPRM relates to full head protection versus half head 
protection. The Safety Board has commented on this ANPRM citing the accident 
at Akron to illustrate the benefits of full head shield and thermal 
protection. 

In its comments, the Safety Board stated that the accident data from the 
past 20 years clearly demonstrate the vulnerability of tank car heads to 
puncture during derailments even, at times, when equipped with shelf-type 
(vertical restraint) couplers. The effectiveness of head shields and thermal 
protection has been equally demonstrated in accidents involving tank cars 
that were so equipped. In addition, the Safety Board cited a recently 
completed study sponsored by the Railway Progress Institute and the AAR 
entitled "Analysis of Tank Cars Damaged in Accidents, 1965 through 1986," 
which concluded that the inclusion of shelf couplers and head shields reduced 
the probability of a head puncture on DOT specification 112 and 114 tank cars 
by 91 percent. The study also noted that 18 percent of the head punctures on 
DOT specification 112, 114, and 105 tank cars during the study period 
occurred in the upper half of the tank head. A second, similarly sponsored 
study entitled "Railroad Tank Car Safety Assessment" concluded that thermal 
protection, head shields, and shelf couplers are "clearly associated with the 
reduced spillage of hazardous materials in recent years." 

The Safety Board's comments also pointed out that following its 
investigation of a collision and derailment at Helena, Montana, in 1989, the 
Board issued Safety Recommendation R-89-80 recommending that the DOT 
evaluate the degree of risk from the release of a hazardous material, 
identify the unacceptable levels of risk, and then modify existing 
regulations to achieve an acceptable level of safety for each product/tank 
car combination. 

The Safety Board further stated that the execution of Safety 
Recommendation R-89-80 would take more than a few months to complete and that 
in the interim, because the need for head shield and thermal protection for 
the transportation of certain products in certain containers has already been 
well established, RSPA should move expeditiously to issue and implement final 
rules that would require full head shields and thermal protection for: 

1. all DOT specification 105 tank cars with a capacity of 18,500 
gallons or less and used to transport flammable gases, 
ethylene oxide, or other products that now require head shield 



67 

and thermal protection when shipped in 105 tank cars exceeding 
18,500 gallons; 47 and 

2. all tank cars transporting class A poisons, materials that are 
toxic by inhalation, and specialty products such as high 
strength acids, chlorine oxidizers, and other extremely 
reactive materials. 

Initial Response to the Derailment by CSX Crewmembers 
and Supervisory Personnel 

The crewmembers of D812-26 testified at the Safety Board's public 
hearing that although they had never been trained on the actions to take 
following an emergency situation, they recognized the importance of 
contacting emergency response personnel immediately following a derailment 
and providing information regarding hazardous materials involved. Their 
onscene actions immediately following the derailment, however, indicate 
otherwise. While the traincrew quickly called and informed the dispatcher of 
the derailment, prudently set up signals to warn approaching trains of the 
derailment, and used their documents to identify the northern- and southern- 
most cars involved in the derailment, there appeared to be no urgency in 
contacting the emergency response personnel onsite and providing the 
necessary information regarding the contents of the tank cars involved in the 
derailment. The front-end crew, apparently believing that either the 
dispatcher or the conductor would provide the necessary information to 
emergency response personnel, were leaving the accident site to get a soft 
drink at a nearby restaurant when they encountered a local police official, 
who then requested that the crewmembers meet with the fire chief. The 
conductor and flagman were preoccupied for more than an hour attempting to 
prevent onlookers from approaching too closely to the burning tank cars and 
never did seek emergency response personnel. While the crew should make 
every effort to protect onlookers from the dangers of derailed tank cars, the 
crew should have also recognized the need to contact emergency response 
personnel when it became evident that emergency response agencies were 
onscene. The Safety Board concludes that the traincrew, contrary to company 
instructions, did not contact as soon as possible emergency response 
personnel onsite to provide them with shipping papers and vital information 
about hazardous materials involved in the derailment. Although the Safety 
Board recognizes the confusion and unpredictable situations that may arise 
following a hazardous materials emergency, the actions of the crew of D812-26 
were not indicative of a crew that had been instructed and trained thoroughly 
about actions to take following an emergency involving hazardous materials. 

Even after the front-end crew came in contact with the fire chief, the 
crew did not convey accurate and complete information to the fire chief 
regarding the location of the other crewmembers and a second copy of the 
consist. After the brakeman and the fire chief returned to the accident site 



All DOT specification 112 and 114 tank cars transporting flammable 
gases and anhydrous ammonia currently must have head shield and thermal 
protection. 



68 

and were unsuccessful in locating the lost profile, the brakeman contacted 
the conductor by radio to let the fire chief talk to the conductor about the 
cars involved in the derailment. However, neither the conductor nor the 
brakeman informed the fire chief that he was talking to the conductor or that 
the conductor and the flagman were at the rear of the train with a second 
copy of the profile (the fire chief believed that he was talking with 
someone in the rail yard). Had the fire chief been informed that he was 
talking to the conductor who was at the rear of the train with a second copy 
of the profile, the fire chief could have sent someone to that location to 
obtain the profile and much of the subsequent skepticism about the cars 
involved in the derailment could have been avoided. The traincrew's failure 
to communicate accurate and complete information to the fire chief again 
suggests a lack of thorough training on the actions to take immediately 
following an emergency involving hazardous materials. 

The actions of the first arriving railroad supervisory personnel suggest 
that first line supervisors also had not been adequately instructed on the 
actions to take immediately following an emergency involving hazardous 
materials. After the CSX trainmaster arrived onscene and talked to the 
flagman by radio, he believed that fire department personnel had been 
provided with the necessary information regarding the derailed tank cars. He 
then returned to Akron Junction. He made no effort to contact the fire chief 
to determine if all necessary information had been provided or to verify the 
accuracy of the information. Although the Safety Board recognizes that 
supervisory personnel may have other responsibilities following a train 
derailment, the Board believes that supervisors must first verify that 
emergency response agencies have received accurate and timely information 
regarding any hazardous materials involved in the derailment. 

The Safety Board has previously addressed the need for traincrews, as 
well as railroad supervisors, to be trained on the actions to be taken 
immediately following a train derailment involving hazardous materials, 
particularly the need to provide emergency response personnel with any 
documentation regarding hazardous materials that may be involved in the 
derailment. In its report of the derailment of a Seaboard Coast Line 
Railroad (now part of CSX Transportation) train at Colonial Heights, 
Virginia, on May 31, 1982, 48 the Safety Board stated: 

Throughout a hazardous materials emergency, and especially during 
the early minutes, it is essential that to the fullest extent 
possible, accurate and complete information be provided to 
emergency response personnel about the hazardous materials which 
present a threat to the safety of the public and the responding 
personnel. How quickly this information is provided to the 
appropriate personnel often determines the magnitude and duration 
of these incidents. The prompt transfer of accurate information is 



48 National Transportation Safety Board. 1983. Derailment of Seaboard 
Coast Line Railroad train No. 120, at Colonial Heights, Virginia, on May 31, 
1982. Railroad Accident Report N T SB / R A R - 83 / 04 . Washington, DC. 45 p. 



69 

one task which the Safety Board has observed repeatedly as being 
the main impediment to an efficient and coordinated response to a 
transportation accident involving hazardous materials. 

As a result of its investigation of that accident, the Safety Board issued 
the following safety recommendations to the railroad: 

R-83-48 

Periodically instruct and test traincrews and supervisory personnel 
on the procedures for using train documents to identify all cars 
transporting hazardous materials and the information to be provided 
to assist emergency response personnel. 

R-83-49 

Require supervisory personnel arriving at the scene of an emergency 
to determine what information has been provided by traincrews to 
emergency response personnel, to verify the accuracy of the 
information provided, and to advise the onscene coordinator of any 
errors or omissions in the initial information given by the 
traincrews. 

The recommendations were classified as "Closed—Acceptable Action" on May 24, 
1983, after the railroad responded that the importance of notifying emergency 
response personnel of any hazardous materials entrained would be stressed in 
rules classes and timetable instructions. 

As a result of its special investigation of the release of oleum during 
wreckage clearing operations after the derailment of a Seaboard System (now 
part of CSX) freight train in Clay, Kentucky, on February 5, 1984, 49 the 
Safety Board, on July 22, 1985, recommended that the railroad: 

R-85-80 

Modify its program of periodic training of train service employees 
to include instructions on the meaning and applications of 
operating rules applicable to an emergency involving hazardous 
materials. 

The recommendation was classified as "Closed—Acceptable Response" on 
March 25, 1986, after the Seaboard System assured the Safety Board that all 
employees would receive instructions on those rules applicable to an 
emergency involving hazardous materials. 



National Transportation Safety Board. 1985. Release of oleum during 

wreckage-clearing following derailment of Seaboard System Railroad train 

Extra 8294 North, Clay, Kentucky, February 5, 1984. Special Investigation 
Report NTSB/S I R-85/01 . Washington, DC. 39 p. 



70 

As a result of its investigation of the derailment of a Baltimore and 
Ohio Railroad Company freight train near Miamisburg, Ohio, on July 8, 1986, 50 
the Safety Board issued the following safety recommendation to CSX: 

R-87-56 

Reemphasize to all operating personnel the importance of directing 
their initial activities following a derailment to the cooperative 
support of local emergency response agencies. 

On March 9, 1988, CSX informed the Safety Board of (1) the bulletins that 
had been issued addressing the "Prevent Accidental Chemical Exposure" 
program, (2) the revised hazardous materials training schedule, and (3) the 
procedures outlined in the CSX's Hazardous Materials Emergency Response 
Guide. Based on the information provided, the recommendation was classified 
as "Closed—Acceptable Action" on July 25, 1988. 

The Safety Board believes that the accident in Akron illustrates that 
CSX personnel were still not provided adequate training on the actions to 
take immediately following an emergency situation despite the Safety Board's 
recommendations on this issue and CSX's assertions that this training was 
being accomplished. Although it appears that CSX management has made the 
necessary information available in the form of bulletins or guidelines, 
operating crews apparently are not understanding or being instructed 
sufficiently on the importance of this information. The Safety Board 
acknowledges CSX's efforts after the Akron accident (1) to provide division 
managers enhanced training on responding to rail transportation emergencies 
involving hazardous materials and (2) to review the feasibility of providing 
video tapes on hazardous materials rules and on emergency response for use in 
operating rules classes. The Safety Board believes, however, that specific 
training on responding to emergencies involving hazardous materials needs to 
be provided to traincrews and supervisory personnel in addition to what is 
covered in operating rules classes for traincrews. This training should 
include, at a minimum, the responsibility of crewmembers to identify 
themselves to emergency response personnel and to provide accurate 
information, including onboard documentation, of hazardous materials 
involved in the accident, and the responsibility of supervisory personnel to 
verify that emergency response personnel have all needed information and that 
it is accurate. The Safety Board also believes that the Association of 
American Railroads should notify its members and reemphasize the need to 
provide traincrews and supervisory personnel adequate training regarding the 
actions to take immediately following a train derailment. 

The investigation revealed that none of the crewmembers had been tested 
on their knowledge of their responsibilities for emergencies involving 
hazardous materials. Further, a CSX trainmaster said that although he did 



50 National Transportation Safety Board. 1987. Hazardous materials 
release following the derailment of Baltimore and Ohio Railroad Company train 
No. SLFR, Miamisburg, Ohio, July 8, 1986. Hazardous Materials Accident 
Report NTSB/HZM-87/01 . Washington, DC. 90 p. 



71 

not evaluate the crewmembers after the accident, he took no exception to 
their actions. The Safety Board believes that evaluation is an integral part 
of employee training; it should be conducted in the classroom and in the 
operating environment to be certain employees understand their 
responsibilities and to provide a measure of the adequacy of the railroad's 
training program. The Safety Board believes that CSX should outline the 
means by which supervisors are to determine if their employees understand 
fully their responsibilities. 

Notwithstanding the Safety Board's concern with the actions of the 
traincrew following the derailment, the Safety Board believes that the 
emergency response personnel should have made every effort to verify the 
location of crewmembers and train documents on both the front-end and rear- 
end of the train. The conductor, who is responsible for the train documents, 
typically rides at the rear of the train when a caboose is used at that 
location. The conductor testified, however, that for the more than 1 1/2 
hours he was at the rear of the train, he never came in contact with any fire 
department personnel. The Safety Board acknowledges that it may be difficult 
to determine when initially arriving at the scene of an accident whether or 
not the train has a caboose. Nevertheless, emergency response personnel 
should routinely inquire about the location of all crewmembers and onboard 
train documents. The Safety Board believes that the International 
Association of Fire Chiefs, in notifying its members of the circumstances of 
the accident at Akron, would be the appropriate organization to emphasize 
this need to local emergency response personnel. 

Because of the problems the fire department experienced in laying hoses 
and obtaining sufficient water supply, the delay in obtaining information 
about the tank cars involved in the derailment did not, in this instance, 
delay the fire department's attack on the fire. However, had there been no 
delay in providing the water supply, insufficient knowledge of the hazardous 
materials involved could have adversely affected the fire department's attack 
on the fire. It is vital that railroad personnel provide, and emergency 
response personnel obtain, as quickly as possible following the accident 
accurate information about the hazardous materials involved in the 
derailment. The Safety Board believes that the breakdown in communicating 
and locating vital information regarding the cars involved in the derailment 
may have resulted, in part, from the lack of jointly conducted emergency 
response drills and exercises between the city agencies and the railroad. 
The fire department and the B.F. Goodrich Chemical Company had engaged in 
emergency drills and preplanning for a disaster, which attests to the well 
organized manner in which the emergency situation at the chemical facility 
was handled. 

The Safety Board has previously addressed the need for local emergency 
response agencies and railroads to conduct joint emergency preparedness 
exercises. The Safety Board's 1985 report 51 on rail yard safety reviewed 



51 National Transportation Safety Board. 1985. Railroad yard safety: 
hazardous materials and emergency preparedness. Special Investigation Report 
NTSB/SIR-85/03. Washington, DC. 25 p. 



72 

the status of emergency preparedness for handling releases of hazardous 
materials in rail yards and concluded that much work remained to be 
accomplished. On April 30, 1985, the Safety Board issued the following 
safety recommendation to all railroads 52 that operate rail yards: 

R-85-53 

In coordination with communities adjacent to your railroad yards, 
develop and implement emergency planning and response procedures 
for handling releases of hazardous materials. These procedures 
should address, at a minimum, initial notification procedures, 
response actions for the safe handling of releases of the various 
types of hazardous materials transported, identification of key 
contact personnel, conduct of emergency drills and exercises, and 
identification of the resources to be provided and the actions to 
be taken by the railroad and the community. 

The Seaboard System Railroad informed the Safety Board on July 21, 1985, that 
it was working with its "partner," the Chessie System Railroads, to ensure an 
appropriate and adequate response to a hazardous materials emergency in a 
railroad yard, including the coordination of its plans and procedures with 
local emergency responders. On February 8, 1988, the CSX informed the 
Safety Board that the Seaboard and the Chessie had merged their operations 
under the name of CSX Transportation, Inc., and this merger included the 
establishment of a single organization of hazardous materials specialists and 
adoption of the best features of both programs. The company informed the 
Safety Board that the CSX Prevent Accidental Chemical Exposure program for 
yards and terminals had been implemented and that this plan included the 
coordination of the emergency response plans and procedures with those in the 
community. 

The Safety Board believes that although the railroad industry in general 
has recognized the need in the past few years to coordinate emergency 
response activity with local authorities and has taken steps toward that 
goal, further efforts by the railroads and the emergency response agencies 
are needed to fully accomplish this goal. This accident illustrates that 
CSX's prior efforts to coordinate emergency response plans and procedures 
with local response agencies had not been fully accomplished with all 
communities through which CSX trains operate. However, after the Akron 
accident, CSX, in cooperation with the fire departments and rescue agencies 
of Summit County and the City of Akron, conducted 6 days of comprehensive 
orientation, training, and field exercises on the handling of hazardous 
materials incidents in rail transportation. CSX indicated that it plans to 
conduct similar training programs at other locations in its system and was 
developing a computerized hazardous materials data program that could provide 
local emergency response agencies information about hazardous materials that 



The Southern Railway System and the Chessie System Railroads were 
excluded as recipients of this recommendation because they already had 
established a corporate policy for meeting the objective of the 
r ec ommenda t i on . 



73 

are being transported through their communities. The Safety Board notes the 
actions taken by CSX after the Akron accident and urges the railroad to 
complete as soon as possible the training program, particularly to conduct 
emergency drills and exercises, with all communities through which CSX trains 
operate. Although Safety Recommendation R-85-53 to the Seaboard System 
Railroad (CSX) was being held in an "Open-Acceptable Action" status based on 
the railroad's efforts to address emergency preparedness, it is now being 
classified as "Closed--Acceptable Action/Superseded" as a result of the new 
recommendation being issued in conjunction with this report on the Akron 
accident. 

Adequacy of Train Consist Information 

There was no Federal or company requirement that after the train 
departed its initial terminal the train consist be updated as cars were 
either added or set off en route. The consist obtained by the Conrail safety 
supervisor in Cleveland and brought to the accident site did not reflect the 
makeup of the train at the time of the derailment. Even though the brakeman 
reviewed this document at the command post and identified which cars had been 
set off and added, the incident commander was not confident that only butane 
was involved in the derailment. As a result, he and the Conrail safety 
supervisor entered the accident area in an attempt to identify those cars 
involved in the derailment and exposed themselves needlessly to hazardous 
conditions. The company required that the conductor prepare a "wheel report" 
listing those cars that had been set off or added en route; however, at the 
time of the accident, the report did not identify the five cars that had been 
added at Warwick. The conductor was able to determine which cars derailed 
and, consequently, which hazardous materials were involved in the derailment 
by compiling information from several documents he carried and from 
information he learned from the front-end crew. 

The onboard train documents are an early source of information for 
emergency response personnel for determining what hazardous materials may be 
involved in a derailment. The Safety Board believes, therefore, that the 
train consist should at all times accurately reflect the location and 
position of hazardous materials cars in the train. Without this up-to-date 
information, emergency response personnel are unable to plan appropriate 
actions. The Safety Board notes that CSX and the railroad industry are 
researching methods of providing aboard each train real-time documentation of 
train consists. The Union Pacific, for example, is experimenting with the 
use of onboard computers to generate real-time consist information. The 
Safety Board believes, however, that until adequate methods are developed, 
the FRA should revise the existing regulations to require that crews update 
train documents when cars are added or set off after the train has departed 
its initial terminal. Until the FRA has acted to revise the regulations, the 
CSX should require this of its operating crews. 



74 

Movement of Damaged Tank Cars 

After the fire department was confident about the information regarding 
hazardous materials in the derailed tank cars, onscene activities were 
accomplished in a timely and professional manner. These activities included 
the response to the tank car fires, the response to the fire at the adjacent 
chemical facility, and the evacuation of residents. The fire department, 
and the city in general, however, depended on the expertise of the railroad 
for the removal of the wreckage from the initial derailment site. The 
operations chief considered it unsafe to unload the product from the tank 
cars at the accident site because of the continuing fire from tank car CITX 
33875 and agreed with CSX's plan to rerail the tank cars and move them to 
Akron Junction yard where the cars would be more permanently secured for the 
movement to Canton--a location with facilities where the product could then 
be offloaded. The railroad, however, did not discuss alternatives with the 
city nor did the railroad advise the city of the possible risks associated 
with rerail ing the tank cars. Only after the second event (when the tank 
car rolled off its trucks while being moved after the derailment) were 
alternative plans and the risks associated with each course of action 
discussed thoroughly with city officials. 

The Safety Board recognizes the limited technical resources that may be 
available to local communities regarding wreckage clearing operations and 
understands the communities' reliance on the railroad to take the appropriate 
course of action. For this reason, it is necessary for the railroad to 
discuss with the local emergency response agencies the severity of known 
damage to tank cars carrying hazardous materials and the dangers posed to 
public safety, all possible courses of actions, and any associated risks. 
However, it is also important that the incident commander, as the person in 
charge of overall activity at the scene of an accident, play an active role 
and search out information about the severity of known tank car damage and 
dangers posed, possible solutions or alternatives, and risks involved. 
Further, the Safety Board is concerned that the presence of the FRA 
personnel, who may have been able to offer technical advice or guidance, was 
not made known to the incident commander until after the second event. 

The decision by the railroad to rerail the tank cars at the initial 
derailment site concerns the Safety Board, aside from the fact that the 
risks associated with such a move were not discussed with local authorities. 
CSX mechanical personnel expressed concern that rerail ing the tank cars for 
further movement was dangerous and not the preferred method. The Safety 
Board concludes that, given the assessment by CSX mechanical personnel and 
the severity of the damage to the tank car body bolsters and couplers, the 
railroad should not have rerailed the tank cars but, rather, should have 
waited until flatcars were available before moving the tanks from the 
derailment site. 



75 

The Safety Board has previously expressed concern about the need for 
written technical guidance to help emergency response personnel assess the 
severity of tank car damage and select the appropriate means to remove the 
wreckage. As a result of the Safety Board's investigation of a freight train 
derailment involving hazardous materials near Inwood, Indiana, on November 8, 
1979, 53 a safety recommendation was issued asking the FRA to: 

1-80-2 

Develop guidelines for handling tank cars containing pressurized 
liquefied gases at accident sites based on research and tests of a 
representative sample of damaged tank cars. 

In its initial response to this recommendation in May 1981, the FRA 
indicated that it agreed with the intent of the recommendation and would 
initiate a program to inspect damaged tank cars. In December 1981, the FRA 
indicated that it would coordinate the program with the AAR. Based on these 
responses, the safety recommendation has been classified as "Open—Acceptable 
Response." On July 3, 1990, the FRA provided an update on 21 open safety 
recommendations related to hazardous materials, including 1-80-2. The FRA 
stated in that letter that the approach to Safety Recommendation 1-80-2 
outlined in 1981 "proved to be impracticable." The FRA, however, provided 
copies of work that has been done on this topic since the issuance of the 
recommendation, including AAR's publication "Tank Car Damage Assessment" and 
an FRA document used in its accident/incident investigation course on 
hazardous materials, portions of which address the assessment of the severity 
of tank car damage. The Safety Board has also been informed that the FRA, in 
conjunction with industry wreckage clearing experts, is currently developing 
technical guidelines to help wreckage clearing personnel assess the severity 
of tank car damage and select wreckage clearing actions based on damage 
assessments. The Safety Board has reviewed the AAR's publication and the 
FRA's investigation course material and believes that these documents and 
FRA's ongoing project with industry wreckage clearing experts address the 
concerns that prompted the Safety Board's recommendation. Pending completion 
of the FRA guidelines for wreckage clearing personnel, Safety Recommendation 
1-80-2 will be classified as "0pen--Acceptable Alternate Response." 

It is important to note that the AAR stated in the preface to its "Tank 
Car Damage Assessment" that these guidelines are to be used by individuals 
who have experience in assessing tank car damage and who are knowledgeable of 
tank car construction requirements and wreckage clearing operations. 
Emergency responders often do not have such technical experience and 
knowledge; however, they should be aware of these guidelines when securing 
the services of someone with this experience and knowledge, when assessing 
the seriousness of dangers posed as a result of tank car damages, and when 
evaluating alternative actions available to minimize those dangers. Further, 
FRA personnel responding to accidents should make their presence and purpose 



National Transportation Safety Board. 1980. Tank car structural 
integrity after derailment. Special Investigation Report NTSB-SIR-80-1. 
Washington, DC. 37 p. 



76 

known to local emergency response personnel, inform local emergency response 
personnel of the guidance currently available for assessing tank car damage 
and wreckage clearing operations, and notify emergency response personnel of 
any imminently hazardous conditions that may exist. 

Location of Chemical Plants and Other Hazardous Materials Facilities 
Near Railroad Tracks 

Had the derailment caused more extensive damage to the B.F. Goodrich 
chemical facility, located adjacent to the railroad tracks, or caused damage 
to the pipelines transporting chemical products at the facility buildings, 
the accident could have been much more severe. The storage and production of 
hazardous materials in close proximity to mainline railroad tracks has long 
been a concern of the Safety Board. 

On March 25, 1981, at Enos, Indiana, a railroad flatcar that had 
derailed struck three of four 1,000-gallon farm truck tanks loaded with 
anhydrous ammonia parked near the mainline tracks. Ammonia escaped from one 
of the breached tanks, mixed with fog, drifted across a divided highway 1/4 
mile away, obscured motorists' vision, and led to multiple motor vehicle 
crashes. The distance from the tanks to the track ranged from about 19 to 40 
feet. The flatcar traveled 65 feet from the track before coming to rest. On 
November 26, 1976, in Belt, Montana, one of several derailed railroad cars 
struck a 16,000-gallon gasoline storage tank. In the ensuing fire, the 
entire bulk storage plant burned; 2 persons were killed and 24 others were 
injured. The tank was located about 42 feet from the mainline track; 
several of the derailed cars traveled more than 100 feet from the track. 

In a study of accidents investigated by the Board from 1976 to 1979, the 
Safety Board found that in 123 of 298 accidents (or about 41 percent), 
derailed cars traveled more than 50 feet (lateral distance) after leaving the 
track. In slightly more than 6 percent of the accidents, cars traveled more 
than 100 feet after leaving the track. 

As a result of these accidents, the Safety Board issued the following 
Safety Recommendations 1-82-1 through -4 to the AAR, 1-82-5 to the National 
Association of Regulatory Utility Commissioners (NARUC), 1-82-6 to the 
National Fire Protection Association (NFPA), and 1-82-7 to the American 
National Standards Institute, Inc. (ANSI): 

1-82-1 

Reevaluate existing practices and standards influencing the 
placement of hazardous materials storage which may be vulnerable to 
damage by derailed railroad cars in train accidents. 



77 

1-82-2 

Based on the results of a reevaluation of existing practices and 
standards, develop necessary changes in recommended practices to 
identify and protect vulnerable hazardous materials storage near 
mainline railroad tracks and disseminate these recommended 
practices to member companies for implementation. 

1-82-3 

In coordination with the National Association of Regulatory Utility 
Commissioners, identify actions States might take to require 
adequate protection of future hazardous materials storage near 
mainline railroad tracks against damage by derailed railroad cars 
in train accidents. 

1-82-4 

Coordinate development of recommended practices for identifying and 
protecting hazardous materials storage near mainline railroad 
tracks with the National Fire Protection Association and the 
American National Standards Institute, to assure consistency among 
related recommended safety practices. 

1-82-5 

Reevaluate State statutes and administrative orders to identify 
action States might take to improve protection of hazardous 
materials storage near railroad right-of-way against damage by 
derailed railroad cars in train accidents, and develop guidelines 
for State actions if needed. 

1-82-6 

Reevaluate National Fire Protection Association No. 30 "Flammable 
and Combustible Liquids Code" to assure adequate protection of 
hazardous materials storage located near mainline railroad tracks 
against derailed railroad cars in train accidents. 

1-82-7 

Reevaluate and amend as necessary American National Standards 
Institute Standard K61. 1-1972, "Safety Requirements for the Storage 
and Handling of Anhydrous Ammonia," to provide adequate protection 
of hazardous materials containers located near mainline railroad 
tracks against derailed railroad cars in train accidents. 

In regard to these safety recommendations, the Safety Board notes the 
efforts of the interindustry task force, established by the Chemical 
Manufacturers Association (CMA) and the AAR to address the safe 
transportation of hazardous materials by rail, and Circular OT-55 



78 

subsequently issued by the AAR to its members on this subject. The Safety 
Board has reviewed the circular and believes that it provides valuable 
guidance on separation distances of hazardous materials from mainline 
railroad tracks. The Safety Board believes, however, that the AAR should 
clarify and emphasize in its circular that hazardous materials storage and 
production facilities (including newly constructed and reconstructed 
facilities, tank cars, cargo tanks, and portable tanks) should be located no 
closer than 100 feet from mainline railroad tracks. The AAR in its recent 
letter of July 25, 1990, addressing Safety Recommendations 1-82-1 through -4, 
and again referencing the work done by the interindustry task force, 
indicated that it intends to work with NARUC, NFPA, and ANSI to encourage 
these organizations to adopt recommendations on storage distances contained 
in Circular OT-55. The Safety Board is aware that the CMA has issued a 
notice to its members urging them to adopt recommendations on storage 
distances contained in the AAR's circular. In view of AAR's efforts, the 
Safety Board believes that the intent of Safety Recommendations 1-82-1 and -2 
has been met and, consequently, these recommendations have been classified 
as "Closed—Acceptable Action." Based on AAR's indication that it will work 
with the NARUC, the NFPA, and the ANSI to coordinate recommended practices 
and proposals with these agencies to assure that proposed separation 
distances are safe and consistent among related standards, Safety 
Recommendations 1-82-3 and -4 will be classified as "Open—Acceptable 
Response" pending the outcome of this joint effort. 

With respect to 1-82-5, the Safety Board noted in 1983 the efforts made 
by NARUC to compile information on State regulations regarding protection of 
hazardous materials stored alongside railroad tracks. However, an analysis 
of this information was never conducted. In view of the increase in the 
transportation of hazardous materials by rail and the possible corresponding 
increase in the storage and production of these materials near railroad 
rights-of-way, the Safety Board believes that the NARUC should again survey 
the States to determine what regulations exist to provide protection of 
hazardous materials stored and produced at these locations and analyze the 
information compiled to determine if additional safeguards are needed. This 
information should then be shared with the AAR to aid in its ongoing efforts 
to coordinate separation distances to ensure that they are safe and 
consistent among related standards. Safety Recommendation 1-82-5 is being 
classified as "Closed—Acceptable Action/Superseded" as a result of the new 
recommendation being issued in conjunction with this report. 

The NFPA responded to Safety Recommendation 1-82-6 on November 15, 1982, 
stating that modification of the existing guidelines was not necessary. The 
recommendation was, consequently, classified as "Closed— Unacceptable Action" 
on February 28, 1983. In view of a 1989 study of over 800 accidents 
conducted by AAR on lateral distances traveled by rail cars after leaving the 
track and in view of AAR's recent statement that it will work with NFPA to 
adopt the recommendations on storage distances, the Safety Board encourages 
the NFPA to revise, as necessary, all codes that address protection of 
hazardous materials storage and production facilities located near mainline 
railroad tracks in accordance with the AAR's study and Circular OT-55. 



79 

A response was never received from ANSI regarding 1-82-7. Because of 
AAR's 1989 study and its efforts to coordinate separation distances, the 
Safety Board is issuing a new recommendation to ANSI to review its standards 
that address protection of hazardous materials storage and production 
facilities near mainline railroad tracks in accordance with the AAR's study 
and Circular OT-55. Consequently, Safety Recommendation 1-82-7 has been 
classified as "Closed—Unacceptable Action/Superseded." 

The Safety Board has previously expressed concern that land-use policies 
by communities do not take into account transportation-related safety issues, 
including the location of railroads or high pressure pipelines near 
populated areas. Because the issues do fall under the jurisdiction of local 
communities, the Safety Board, as a result of its investigation of a train 
derailment and subsequent pipeline rupture in San Bernardino, California, 54 
issued safety recommendations to the National Association of Counties and the 
National League of Cities urging them to inform their members of the land-use 
guidance for enhancing public safety contained in the National Research 
Council's Special Report 219, "Pipeline and Public Safety," and encouraging 
them to develop and implement policies to protect public safety for lands 
adjacent to pipelines and railroads. As the Safety Board stated in its 
report of the San Bernardino accidents: 

The recommended actions in this report are specifically directed 
to public safety and land-use issues for pipelines, but the Safety 
Board believes, in principle, the discussion on land use would 
apply to railroads. Moreover, many of the considerations on land- 
use limitations for property adjacent to pipelines but not yet 
developed, also should be applied to land adjacent to railroads 
that has not been developed. 

Similarly, the Safety Board believes that public safety policies on the 
uses of land adjacent to railroad tracks, including the location of hazardous 
materials storage and production facilities and the products being 
transported by rail, should be developed and implemented and urges the city 
of Akron to do so. Further, the Safety Board believes that the National 
League of Cities and the National Association of Counties should inform their 
members of the circumstances of the accident in Akron, and urge members, when 
developing or revising land-use policies on the location of hazardous 
materials storage and production facilities near railroad tracks, to require, 
as a minimum, setback distances consistent with the Association of American 
Railroads' (AAR) study on lateral distances traveled by derailed rail cars 
and AAR's Circular OT-55 "Recommended Railroad Operating Practices for 
Transportation of Hazardous Materials." 



54 Railroad Accident Report NTSB/RAR-90/02, 



80 
CONCLUSIONS 



Findings 

1. The heavy rail burn marks on the B-end left sideframe of WSOR 501003 and 
the lack of similar marks on other cars indicate that WSOR 501003, the 
13th car in the train behind the locomotives, was the first car to 
derail . 

2. The excessive gib clearance on the B-end truck on WSOR 501003 and out- 
of-limits side bearing clearances existed before the accident and 
should have been detected by Northern Rail Car Corporation and railroad 
car inspectors. 

3. Designated inspectors failed to detect excessive gib clearances on WSOR 
501003 because these inspectors apparently have a tendency to look for 
damaged or excessively worn parts rather than mismatched bolsters or 
sideframes; this suggests that the training of the designated inspectors 
or the procedure for inspection of gib clearance was inadequate. 

4. The marginal condition of the track in the area approaching the point of 
derailment and the existence of two turnouts in a curve caused 
increased lateral and vertical movement of car WSOR 501003 as it moved 
over the tracks. 

5. The track was in marginal condition because track rehabilitation had not 
been performed for more than a year. 

6. The quality and thoroughness of the Federal Railroad Administration's 
inspection of the track before the accident was inadequate. 

7. The investigation was hampered in its effort to accurately determine the 
manner in which the train was being operated as it approached the 
accident site because the locomotive units were only equipped with 
three-event recorders. 

8. Inadequate quality control procedures at the Northern Rail Car 
Corporation facility at the time the WSOR series cars were being rebuilt 
resulted in undetected deficiencies. 

9. The Association of American Railroads (AAR) could not be certain that 
deficiencies noted during a routine AAR inspection of the Northern Rail 
Car Corporation facility in January 1987 had been corrected because no 
followup was conducted. 

10. The Association of American Railroads' inspection of Northern Rail Car 
Corporation's rebuild project was deficient; further, inspection of a 
single car was insufficient to determine if all cars in the project 
were being rebuilt to the appropriate specifications. 



81 

11. The failure of the traincrew to immediately contact and provide 
emergency response personnel with shipping papers and vital information 
about hazardous materials involved in the derailment and the failure of 
first arriving railroad supervisory personnel to verify that all 
necessary information had been provided to emergency response personnel 
suggests that crews and first-line supervisors had not been instructed 
and trained properly on actions to take immediately following an 
emergency involving hazardous materials. 

12. Emergency response personnel failed to verify the location of all 
crewmembers and train documents on both the front-end and rear-end of 
the train. 

13. Inadequate communications between railroad and emergency response 
personnel about vital information regarding the cars and hazardous 
materials involved in the derailment may have resulted, in part, from 
the lack of jointly conducted emergency response drills and exercises 
between the city agencies and the railroad. 

14. The onboard documents to which emergency response personnel initially 
refer may not accurately reflect the position of hazardous materials in 
the train because there is no requirement that these documents be 
updated as cars are set off or added en route. 

15. The railroad and the incident commander did not adequately communicate 
about the possible alternatives to, and risks associated with, moving 
the damaged cars from the initial derailment site. 

16. The railroad's decision to rerail the tank cars for further movement was 
not prudent in view of the assessment by CSX mechanical personnel and 
the severity of damage to the bolsters and couplers. 

17. The location of the B.F. Goodrich facility adjacent to the mainline 
railroad tracks could have increased the severity of the accident 
because of the chemicals used within the facility. 



Probable Cause 

The National Transportation Safety Board determined that the probable 
cause of the derailment of train D812-26 in Akron, Ohio, on February 26, 
1989, was the inadequate rebuild and quality control procedures of the 
Northern Rail Car Corporation car repair facility and the inadequate 
inspections of car WSOR 501003 by designated car inspectors that permitted 
the car to enter and continue in service with excessive gib clearance and 
out-of-1 imits side bearing clearance. Contributing to the accident was the 
marginal condition of the track as a result of Conrail's decision to delay 
rehabilitation of the track or not to place a slow order on the track. 



82 

RECOMMENDATIONS 

As a result of its investigation of this accident, the National 
Transportation Safety Board made the following safety recommendations: 

--to CSX Transportation: 

Provide training, in addition to operating rules classes, to 
operating crews and supervisors on the actions they are to take 
immediately following an accident involving hazardous materials; 
this training should include, at a minimum, (1) the responsibility 
of crewmembers to identify themselves to emergency response 
personnel and to provide accurate information, including onboard 
documentation, of hazardous materials involved in the accident, (2) 
the responsibility of supervisory personnel to verify that 
emergency response personnel have all needed information and that 
it is accurate, and (3) the means by which supervisors are to 
determine if employees understand fully their responsibilities. 
(Class II, Priority Action) (R-90-28) 

Complete, as soon as possible, drills and exercises for handling 
releases of hazardous materials with all communities through which 
CSX hazardous materials trains operate. (Class II, Priority 
Action) (R-90-29) 

Require supervisory personnel to explain, before implementing 
wreckage clearing activities, to local emergency response agencies 
the alternative actions considered as well as the planned action 
and the risks associated with each. (Class II, Priority Action) 
(R-90-30) 

Require traincrews to update the train consist to accurately 
reflect the position of hazardous materials as cars are added or 
set off en route. (Class II, Priority Action) (R-90-31) 

--to the City of Akron: 

Develop and implement public safety policies on the uses of lands 
adjacent to railroad tracks, including the location of hazardous 
materials storage and production facilities and the products being 
transported by rail. (Class II, Priority Action) (R-90-32) 

--to the Association of American Railroads: 

Notify members of the Association of American Railroads about the 
circumstances of the accident in Akron, Ohio, on February 26, 1989, 
emphasizing the need (1) for car inspectors to check side bearing 
and gib clearances during inspections, and (2) for traincrews and 
supervisory personnel to be provided adequate training regarding 
the actions to take immediately following a train derailment. 
(Class II, Priority Action) (R-90-33) 



83 

Require followup action on deficiencies noted during inspections of 
freight car repair facilities to determine if the deficiencies have 
been corrected. (Class II, Priority Action) (R-90-34) 

Develop and implement procedures that provide for thorough and 
adequate initial and followup inspections of cars during a rebuild 
project. (Class II, Priority Action) (R-90-35) 

Clarify and emphasize in Circular OT-55 that hazardous materials 
storage and production facilities (including newly constructed and 
reconstructed facilities, tank cars, cargo tanks, and portable 
tanks) should be located no closer than 100 feet from mainline 
railroad tracks. (Class II, Priority Action) (R-90-36) 

--to the Federal Railroad Administration: 

Evaluate the adequacy of track inspections being conducted on the 
Akron subdivision by Federal Railroad Administration inspectors and 
institute necessary changes to ensure thorough and consistent 
inspections. (Class II, Priority Action) (R-90-37) 

Revise 49 CFR 174.26(b) to require the traincrew to maintain, at 
all times, a document reflecting the current position of hazardous 
materials cars in the train. (Class II, Priority Action) (R-90-38) 

Require that Federal Railroad Administration personnel responding 
to a derailment involving hazardous materials (1) make their 
presence and purpose known to local emergency response personnel, 
(2) advise local authorities of guidance available for assessing 
tank car damage and wreckage clearing operations, and (3) notify 
emergency response personnel of any imminently hazardous conditions 
that may exist. (Class II, Priority Action) (R-90-39) 

--to the International Association of Fire Chiefs: 

Notify members of the International Association of Fire Chiefs 
about the circumstances of the accident in Akron, Ohio, on February 
26, 1989, and urge them to emphasize to the appropriate emergency 
response personnel the need (1) to locate all crewmembers and train 
documents as a priority action when responding to a train accident, 
and (2) to search out information about the severity of known tank 
car damage and dangers posed, possible solutions or alternatives, 
and risks involved. (Class II, Priority Action) (R-90-40) 



84 

-to the National League of Cities and the National Association of Counties: 

Inform your members of the circumstances of the CSX derailment in 
Akron, Ohio, on February 26, 1989, and urge them, when developing 
or revising land-use policies on the location of hazardous 
materials storage and production facilities near railroad tracks, 
to require, as a minimum, setback distances consistent with the 
Association of American Railroads' (AAR) 1989 study on lateral 
distances traveled by derailed rail cars and AAR's Circular No. 
OT-55, "Recommended Railroad Operating Practices for Transportation 
of Hazardous Materials." (Class II, Priority Action) (R-90-41) 

-to the National Fire Protection Association: 

Revise, as necessary, the Association's codes that address 
protection of hazardous materials storage and production facilities 
located near mainline railroad tracks, in accordance with the 
Association of American Railroads' (AAR) 1989 study on lateral 
distances traveled by derailed rail cars and AAR's Circular No. 
OT-55, "Recommended Railroad Operating Practices for Transportation 
of Hazardous Materials." (Class II, Priority Action) (R-90-42) 

--to the American National Standards Institute, Inc.: 

Revise, as necessary, the Institute's standards that address 
protection of hazardous materials storage and production facilities 
located near mainline railroad tracks, in accordance with the 
Association of American Railroads' (AAR) 1989 study on lateral 
distances traveled by derailed rail cars and AAR's Circular No. 
OT-55, "Recommended Railroad Operating Practices for Transportation 
of Hazardous Materials." (Class II, Priority Action) (R-90-43) 

--to the National Association of Regulatory Utility Commissioners: 

Survey State regulations to determine if additional safeguards are 

needed to provide protection of hazardous materials storage and 

production facilities near railroad rights-of-way and inform the 

Association of American Railroads (AAR) to aid the AAR in its 

efforts to coordinate separation distances between storage and 

production facilities and mainline railroad tracks. (Class II, 
Priority Action) (R-90-44) 



85 
BY THE NATIONAL TRANSPORTATION SAFETY BOARD 



JAMES L. KOLSTAD 

Chairman 

SUSAN N. COUGHLIN 

Chairman 

JOHN K. LAUBER 
Member 

JIM BURNETT 

Member 

CHRISTOPHER A. HART 

Member 



Adopted: September 25, 1990 



87 
APPENDIXES 
APPENDIX A 
INVESTIGATION AND HEARING 

Investigation 

The National Transportation Safety Board was notified of the accident 
about 8:30 p.m. eastern standard time on February 26, 1989. An investigative 
team was dispatched from headquarters in Washington, D.C., and from field 
offices in Chicago, Illinois; Denver, Colorado; and Ft. Worth, Texas. The 
team arrived at the accident scene on February 27, 1989. Investigative 
groups were established for operations, track, mechanical, hazardous 
materials, and emergency response factors. 

Hearing 

As part of its investigation of this accident, the Safety Board 
conducted a public hearing June 26-29, 1989, in Cleveland, Ohio. Parties to 
the investigation included the City of Akron, State of Ohio, CSX 
Transportation, Inc., Consolidated Rail Corporation, Federal Railroad 
Administration, Bureau of Explosives of the Association of American 
Railroads, B.F. Goodrich Chemical Company, and Northern Rail Car Corporation. 
Additionally, the Brotherhood of Maintenance of Way Employees and the 
Brotherhood of Railway Carmen were parties to the public hearing. Testimony 
was taken from 23 witnesses. 



89 

APPENDIX B 

DEPARTMENT OF TRANSPORTATION 
EMERGENCY RESPONSE GUIDE 22 



Source: U.S. Department of Transportation, Research and Special 
Programs Administration, Office of Hazardous Materials 
Transportation. 1987. Emergency response guidebook: 
guidebook for initial response to hazardous materials 
incidents. DOT P 5800.4. Washington, DC. 76 guides plus 
supplemental pages. 




POTENTIAL HAZARDS 

FIRE OR EXPLOSION 

Extremely flammable; may be ignited by heat, sparks or flames. 
Vapors may travel to a source of ignition and flash back. 
Container may explode in heat of fire. 
Vapor explosion hazard indoors, outdoors or in sewers. 
HEALTH HAZARDS 

Vapors may cause dizziness or suffocation. 

Contact will cause severe frostbite. 

Fire may produce irritating or poisonous gases. 

EMERGENCY ACTION 

Keep unnecessary people away; isolate hazard area and deny entry. 

Stay upwind, out of low areas, and ventilate closed spaces before entering. 

Self-contained breathing apparatus (SCBA) and structural firefighter's protective 

clothing will provide limited protection. 
Isolate for 1/2 mile in all directions if tank car or truck is involved in fire. 
CALL CHEMTREC AT 1-800^24-9300 AS SOON AS POSSIBLE, especially if there is 

no local hazardous materials team available. 
FIRE 

Let tank car, tank truck or storage tank burn unless leak can be stopped; with 

smaller tanks or cylinders, extinguish/isolate from other flammables. 
Small Fires: Dry chemical, C02 or Halon. 
Large Fires: Water spray or fog. 

Move container from fire area if you can do it without risk. 
Cool containers that are exposed to flames with water from the side until well after 

fire is out. Stay away from ends of tanks. 
For massive fire in cargo area, use unmanned hose holder or monitor nozzles; if 

this is impossible, withdraw from area and let fire burn. 
Withdraw immediately in case of rising sound from venting safety device or any 

discoloration of tank due to Tire. 
SPILL OR LEAK 

Shut off ignition sources; no flares, smoking or flames in hazard area. 
Do not touch spilled material; stop leak if you can do it without risk. 
Use water spray to reduce vapors; isolate area until gas has dispersed. 
FIRST AID 

Move victim to fresh air and call emergency medical care; if not breathing, give ar- 
tificial respiration; if breathing is difficult, give oxygen. 
In case of frostbite, thaw frosted parts with water. 
Keep victim quiet and maintain normal body temperature 



91 
APPENDIX C 
HAZARDOUS MATERIALS INFORMATION 



Source: U.S. Department of Transportation, U.S. Coast Guard. 
1984. CHRIS [chemical hazard response information system] 
hazardous chemical data. Commandant Instruction 
M16465.12A. Washington, DC. Vol. 2. 



92 



ACRYLONITRILE 



ACN 



Vinyl cyanide 



Colorless to light tmtatmg oc 

Poisonous, flammable vapor re produced 



Water 

Pollution 



r BE PRODUCED IN FIRE 



t explode rt ^nrted » 



. TO AQUATIC LIFE IN VERY LOW CONCENTRATIONS 



May be dangerous rf 



1. RESPONSE TO DISCHARGE 
(See Response Matnodi Handbook) 



3. CHEMICAL DESIGNATIONS 
3.1 CO Compatibility Claw: Subsumed • 
JJ Formuto: CH, =CHCN 
3.3 MO/UN O^tonatton: 3 1/1093 
3 4 DOT ID NO.: 1093 
3.6 CAS Registry No. 107-13-1 



4. OBSERVABLE CHARACTERISTICS 
4.1 Physical Stata (aa ehfpped): I-JQuk) 
4 J Cotor Colorless 
4.3 Odor MM. pungent, resembling thai of 



Personal Protective Equlpmant Air-s 



HEALTH HAZARDS 

jppiied mask, industrial chemical type, with approved 

v lo mow o' hydrogen cyanide Vapor nhalabon may 

J, abdominal pain, and vomiting Similar symptoms aha 

through the skin, lesser amounts causa stinging and 

; causes severe imtat>on ingestion produces nausea. 

vomiting and abdominal pain 

Treatment of Exposure Skilled medical treatment 

exposure INHALATION remove victim to fresh a 

when entenng contaminated area ) INGESTION i 



a thoroughly v. 



while doing this Replace ampule when its 

Threshold Umit Value: 2 ppm 

Short Term inhalation umitt 40 ppm lor 3 

Toilcttv by Ingestion: Grade a LD»o SO to 

Late Toxictry Data noi available 

Vapor (Oaa) Irritant Characteristics: v arc 

LiquM or Solid Irritant Characteristics: it a 

smarting and iedd*ning of the skin Large 



EYES I 



' (Wear an oxygen or fresh -air- supphed 
■dues vomrtjng by admmralenng strong s 



» breathing, grve artificial 



l guinea c-gi 

irritating such t 

clothing and allowec 



mng 






6. FIRE HAZARDS 




10. HAZARD ASSESSMENT CODE 


6.1 Flash Point 30'F C.C ... 31*F OC. 


(Bee Haxard Assessment Handbook) 


6J FssmmaMa Lirnlts In Air 3 05%-17.0% 




A-P-Q-B-S-Z 


6.3 Ftre Extinguishing Agents. Dry cherrscal. 






alcohol loam, carbon dioxide 






6.4 Fir* Extinguishing Agents Not to be 










Used: Water or foam may cause frothing 




11 HAZARD CLASSIFICATIONS 


6.5 Special Hazards of Combustion 






Products: When heated « burned. ACN 


11.1 


Code df Federal Regutotoona: 


may evofve toxic hydrogen cyarade gas 






and oxides Of nitrogen 


11J 


HAS Hazard Rating for Bulk Water 






Transportation: 


and may travel a considerable drslance to 




Category Rating 


a source of ignrbon and Hash back. May 




Flfe 3 


potymenze and explode 














6.6 Electrical Haxard: Class I. Group D 








6.6 Burning Rale: Data noi available 






6.10 Adtabattc Flame Temperature: 






Data noi available 




Human Toxicity . 3 


(Continued) 




AquabC Toxtcrty 4 
Aesthetic Eftect 3 




7. CHEMICAL REACTIVITY 




Reactrvrty 








7.2 naacthrtty with Common Uatertaia. 






Penetrates leather, so contaminated 


11 J 


NFPA Hazard Claaalnca Bon 




Category Classification 






destroyed Attacks alurrsnum m hrgh 




Reactivity (Yellow) 2 


7 J StaMttty During Transport Stable 












Caustics. Not pertjnenl 






7 .5 Pen/meftrsbon Msy occur spontaneously 






m absence Of oxygen or on exposure to 






vtsrtMe light or excessrve heat wMentty 
si the presence of alkali. Pure ACN is 
















pressure development. The commercial 


12.1 


Physical State at 15'C and 1 eon: 


product is nrkbrted and not subject to 




Liquid 


thrs reaction 


12.2 


Molecular Weight: S3 06 




12J 


Boiling Point at 1 etm. 

17TF = 77 4-C = 350 6-K 


(Continued) 


12.4 


Freezing Point: 
















81 Aquatic Toxtctty: 




505*F = 263X = 536'K 


100 ppm/24 hr/aJi hah/100% killed/fresh water 


12.6 


Critical Pressure: 


OOS-t ppm/24 hr/bkssgn /lethal /sari water 




660 paja = 45 aim = 4 6 MN/m> 


6.2 wateriowi Toxicity Not pertinent 


12.7 


Specrttc Qravhy 


«3 BtotogteaJ Oxygen Oamand (BOO): 




B07S at 20'C (hQu-d) 


70%. S days 


12.6 


UquM Surface Tenalon: Not pertinent 


1.4 Food Chasn Concentration Potential: 


12.9 


liquid Water Intertactol Tension: 


None noted 


12.10 


Vapor (Oaa) SpechV Oravtty: 1 6 




12.11 


Ratio of Specific Hears of Vapor (Oaa): 




12.12 
12.13 


Latent Heat of Vaporization: 
265 Btu/tb = 147 cai/g = 
616 X 10» J/kg 

Heat of Combustion — 14.300 Blu/lb 
= —7930 cal/g = 332 X 10' J/kg 


9. SHIPPING INFORMATION 


12.15 


Heat of Potymertzatton: Noi penineni 


8.1 Grade* of Purity: Technical 98-100% 


12.2! 


Heat of Fusion Data not available 


0.2 Storage Temperature: Ambient 


MM 


Urnlttng Value: Data not available 




12.27 


Retd Vapor Preeaure: 3 5 psia 


t.4 Venting: Pressure-vacuum 






6. FIRE HA7J 


RDS (Continued) 


8.11 StolchlometTlc Air to Fuel Ratio: Data not 


vailaWe 




6 12 Flame Temperature: Data not available 






7. CHEMICAL RE, 


ACTIVITY (Continued) 


7.6 Inhsbttor of Poh/mertiapon: Memythydroqum 


x>e (35-45 ppm) 


7.7 Motor Ratio (Reactant to Product): Data no 






7.8 Reactivity Group IS 







93 




12.17 


12.16 


12.19 


12.20 


SATURATED LIQUID DENSITY 


LIQUID HEAT CAPACITY 


LIQUID THERMAL CONDUCTIVITY 


LIQUID VISCOSITY 


Temperature 
(degrees F) 


Pounds per cubic 
foot 


Temperature 
(degrees F) 


British thermal unit 
per pound-F 
(estimate) 


Temperature 
(degrees F) 


British thermal 
unit-inch per hour- 
square toot-F 


Temperature 
(degrees F) 


Centi poise 





52.800 


28 


.499 


75 


1.150 




N 


5 


52.620 


30 


.499 


80 


1.143 




O 


10 


52.450 


32 


.499 


85 


1.136 




T 


15 


52.280 


34 


.499 


90 


1.128 






20 


52.100 


36 


.499 


95 


1.121 




P 


25 


51.930 


38 


.499 


100 


1.114 




E 


30 


51.760 


40 


.499 


105 


1.107 




R 


35 


51.580 


42 


.499 


110 


1.099 




T 


40 


51.410 


44 


.499 


115 


1.092 




I 


45 


51.240 


46 


.499 


120 


1.085 




N 


50 


51 060 


48 


.499 


125 


1.078 




E 


55 


50.890 


50 


.499 


130 


1.070 




N 


60 


50.720 


52 


.499 


135 


1.063 




T 


65 


50.540 


54 


.499 


140 


1.056 






70 


50.370 


56 


499 


145 


1.049 






75 


50 190 


58 


.499 


150 


1.041 






eo 


50.020 


60 


.499 


155 


1.034 






85 


49.850 


62 


.499 


160 


1.027 






90 


49.670 


64 


.499 










95 


49.500 


66 


.499 










100 


49.330 


68 


.499 










105 


49.150 


70 


.499 










110 


48.980 


72 


.499 










115 


48.810 


74 


.499 










120 


48.630 


76 


.499 










125 


48.460 


78 


.499 











12.21 
SOLUBILITY IN WATER 


12.22 
SATURATED VAPOR PRESSURE 


12.23 
SATURATED VAPOR DENSITY 


12.24 
IDEAL GAS HEAT CAPACITY 


Temperature Pounds per 100 
(degrees F) pounds ol water 


Temperature 
(degrees F) 


Pounds per square 
inch 


Temperature 
(degrees F) 


Pounds per cubic 
foot 


Temperature 
(degrees F) 


British thermal unit 
per pound-F 


70.02 8000 





.193 





.00208 





.261 




10 


.277 


10 


.00291 


25 


.270 






20 


.390 


20 


.00402 


50 


.280 






30 


.540 


30 


.00545 


75 


.289 






40 


.735 


40 


.00727 


100 


.297 






50 


.987 


50 


.00957 


125 


.306 






60 


1.306 


60 


.01242 


150 


.314 






70 


1.707 


70 


.01593 


175 


.323 






80 


2.205 


80 


.02019 


200 


.331 






90 


2.815 


90 


.02532 


225 


.338 






100 


3.558 


100 


.03142 


250 


.346 






110 


4.452 


110 


.03863 


275 


.354 






120 


5.520 


120 


.04707 


300 


.361 






130 


6.786 


130 


.05688 


325 


.368 






140 


8.274 


140 


06820 


350 


.375 






150 


10.010 


150 


.08117 


375 


.382 






160 


12.030 


160 


.09594 


400 


.389 






170 


14.350 


170 


.11260 


425 


.395 






180 


17.010 


180 


.13150 


450 


.401 






190 


20.040 


190 


.15250 


475 


.408 






200 


23 480 


200 


.17590 


500 


.414 






210 


27.360 


210 


.20190 


525 


.420 






220 


31.710 


220 


.23060 


550 


.425 






230 


36.570 


230 


.26210 


575 


.431 












600 


.437 



94 



AMMONIA, anhydrous 



AMA 



Common Synonyms 


Liquefied compressed Colorless Ammonrt odor 


US-JAn"™-". 


gas 




Floats and boils on water Poisonous, vtaiole vapor doud rs produced 


*ea- gogoti %. ■ :-■= : :•*?--; Kz.a'i~i a~5 -uDBe- overcoming 


S:a. -p».rc'a'- -s. *a^ i-a. ■■; .no;, oo*- vaoc- 


'SO V-tV "V---^ ,,-j.j,- ~ai*-a 


N-', ;:a -*<•• ,-::: -• -:•.- ag*-;.^ 




Combustible 


Fire 


: -, : .. ■ a "t-s anc protect men eHectingshuto'' wtr> water 




VAPOfl 




POISONOUS IF INHALED. 




^muting to^eyes. nose a oa 




LIQUID 


Exposure 


Ha'nXTrilwaMrSed 85 




_W,li cause (rosttnte . „- : -• , t . 




'■■' •.---- - "- -!-" - : ;\-W : -- ." w.r pift o' watei 








HARMFUL TO AQUATIC LIFE IN VERV LOW CONCENTRATIONS. 


Water 


May be dangerous rt il enters waier mtekes 


Pollution 


'■ ■•->••* -■ — ■'■ — ■-« 


1. RESPONSE TO DISCHARGE 


2 LABEL 


(Sw R«*ponM Uamoda Handbook) 


2.1 Category Nonflammable gas 




2.2 Class : 






to "™'""" S " 




3. CHEMICAL DESIGNATIONS 


4 OBSERVABLE CHARACTERISTICS 


3. i CG Compatibility Class: Ammonia 


4.1 Physic* State is* anipped.. 


3.2 Formula: NH 3 


Compressed hquefwd gas 


3.3 IWO/UN Designation: 20/1006 


4.2 Color Cotortess 


3.4 DOT ID HO.. 1005 


4.3 Odor Pungent emremety pungent 


3.5 CAS Registry No.. 7664-41-7 




5. HEALTH HAZARDS 


5.1 Personal Protective Equipment: Gas-tigm chemical goggles Berf-coniamed breaming apparatus. 


rubber boots, rubber gloves, emergency shower and eye bath 


5.2 Symptom* Following Eiposure: 700 ppm causes eye imtaion. and permanent m|ury may result 


inflammation or edema of the larym Contact ot me liquid with Hun treeies the ttsaue and men 


5.3 Treatment of Exposure: INHALATION move v<ctim to fresh air and give artificial respwabon if 


necessar> Oxygen may be uselul Observe lor laryngeal spasm and perlorm tracheostomy rt 


indicated SkiN OR EVES flood immediately with running water lor 15 mm Treat suOaeoueniiy 


5.4 ThrtanoM Umlt Value: 25 ppm 


5.5 Snort Term Inhalation Umlta: 50 ppm lor 5 mm 


5.8 Toirdty by Ingestion: Not pertinent 


5.7 late Toilctty: Not pertinent 


5.6 Vspor (Gss) trmsnt Char act aria ties ; Vapors cause severe eye or itvoat -miai-on and may cause 




5 9 Liquid or Solid Irritant Char act eristics Causes smarting ol the Skm and first-degree bums on 




5 10 Odor Threshold: 46 6 ppm 


5.11 IDLH Value: 500 ppm 



6 FIRE HAZARDS 


10. HAZARD ASSESSMENT CODE 


6.i Flesh Point Not ftammaole under 


(See Hazard Assessment Handbook) 


conditions likely to be encountered 


A-3-C-K-L-M-n-O 


6-2 Flafwna*at Umrta m Air 15 50*27 00% 




8.3 Fire Eittngvlshrng Agents: Stop How of 








8.4 Fire ErtmgvtsWng Agents Not to be 


11. HAZARD CLASSIFICATIONS 


Usad: None 

6.5 Sped*] Hazards of Comoustton 

6.6 Senavfor m Fire: Not pertinent 

8.7 Ignreon Temperature: 1204'F 

8.8 Electrical Hazard: Class 1, Group D 


11.1 Code of Federal Regulations: 

Nonflammable gas 


11.2 NAS Hazard Rating for Bulk Water 
Transportation: 

Category Rating 


8.9 Burning Rate: 1 mm/mm. 


Frre t 


6.10 Adsabaoc Flame Temperature: 


Health 


6.1 1 SWchtometrtc Air to Fuel Ratio: 

6050 (Est) 


Lxjutd ot Solid Imiant . . 2 


8.12 Ftsme Temperature: Oata not available 






Aquatic Toxicity 2 

Aesthetic Effect 2 




7. CHEMICAL RtACTIVITT 


Reactivity 


7.1 Reecttvtty WKh Water Dissolves with mW 


Other Chemicals 2 


heal effect 
7 2 ReectMty srfth Common Materials 

Corrosive to copper and galvanaed 


11.3 NFPA Hazard Classification: 

Category Classification 


7 J StabWty During Transport: Stable 
7.4 Neutralizing Agents for Adds and 
Caustics. Dilute with water 


Flammability (Red) 1 i 
Reactivity (Yellow) C 


7.5 Poh/mertzabon: Not pertmeni 




7.6 tfMbriOf ot Porymertaaflon: 




Not BOrOnenl 




7.7 fttottr Rstw (Reactant to 




Product): Oata not available 




7 8 Reectivtty Oroup: Data not available 






12. PHYSICAL AND CHEMICAL PROPERTIES 




12 i Physical State at 16'C and 1 ami- 




Gas 




12.2 Molecular Weight: 17 03 




12.3 Boiling Point at 1 ttm: 




-28 TF = — 33.4-C = 239 6*K 




— 10B-F = —77 7*C = 265 S'K 
12.5 CrtfJcal Temperature: 


6. WATER POLLUTION 


8i Aquatic Toiletry-. 


271 "F = 133-C = 406-K 


2 ■ 2.5 ppm/ 1-4 aays/gokjhsh and yellow 


12.6 Critical Pressure; 


perch/LC 


1636 psia = til 3 aim = 


60 ■ 80 ppm 3 Oays/crayfrsh/LCioo 


11.27 MN/m, 


6 2 ppm/ 96 hr/ fathead minnow /TL. 


12.7 Specific Gravity: 


8.3 Waterfowl Toilctty: 120 ppm 


682 at -33 4-C (liquid) 


8.3 BsWOQteal Oxygen Demand (BOO): 


12.6 Liquid Surface Tension: Not pertmeni 
12.9 Liquid Water Intartaclal Tension: 


6.4 Food Chain Concentration Potential: 






12.10 Vapor (Gas) Specific Gravity 6 




12.11 Ratio of Speetflc Heats of Vapor (Oaa): 




13 ai 20*C 








586 Btu/lb = 327 cal/g = 




13 7 X 10' J/kg 




12.13 Heal ot Combuatton: —7992 Btu/ib 




*440 cai/g 185 9 X 10" j/kg 

12.14 Heel or Decompoaroon: Not pertinent 

12.15 Heat of Solution: -232 Btu/lb 


9. SHIPPING INFORMATION 


91 Grades of Purity: Commercial, industrial 


= —129 cal/g = -5 40 X 10 s J/kg 


retngerahon, electronic, and 


12.16 Heat ot Polymerization: Not pertinent 


metallurgical grades all have punty 


12.25 Heat of Fualon: Date not available 


greater than 99 5% 


12.26 Limiting value: Data not available 


6.2 Storage Temperature: Ambient for 


12.27 field Vapor Pressure: 21 1 9 psia 


pressumed ammonia, low temperature 




tor ammonuj at atmosphenc pressure 




9.3 Inert Atmosphere: No requirement 




9.4 Venting: Saiety retiel 250 psi lor ammonia 




NOTES 



95 




12.17 


12.18 


12.19 




SATURATED LIQUID DENSITY 


LIQUID HEA 


T CAPACITY 


LIQUID THERMAL CONDUCTIVITY 


LIQUID VISCOSITY 


Temperature 


Pounds per cubic 


Temperature 


British thermal unit 


Temperature 


British thermal 
unit-inch per hour- 
square toot-F 






(degrees F) 


foot 


(degrees F) 


per pound-F 


(degrees F) 


(degrees F) 




—105 


42.070 


—75 


1.041 




N 




N 


—100 


42.200 


—70 


1.043 









O 


—95 


42.310 


—65 


1.046 




T 




T 


—90 


42.410 


—60 


1.049 










—85 


42.500 


—55 


1.052 




P 




P 


—80 


42.570 


—50 


1.054 




E 




E 


—75 


42.630 


—45 


1.057 




R 




R 


—70 


42.680 


—40 


1.060 




T 




T 


—65 


42.720 


—35 


1.063 




1 




1 


—60 


42.740 


—30 


1.066 




N 




N 


—55 


42.750 








E 




E 


—50 


42.750 








N 




N 


—45 


42.730 








T 




T 


—40 


42.700 














-35 


42.660 














—30 


42.600 















12.21 
SOLUBILITY IN WATER 


12.22 
SATURATED VAPOR PRESSURE 


12.23 
SATURATED VAPOR DENSITY 


12.24 
IDEAL GAS HEAT CAPACITY 


Temperature 
(degrees F) 


Pounds per 100 
pounds of water 


Temperature 
(degrees F) 


Pounds per square 
inch 


Temperature 
(degrees F) 


Pounds per cubic 
foot 


Temperature 
(degrees F) 


British thermal unit 
per pound-F 




M 


—40 


10.470 


—40 


.03957 





.487 




I 


—35 


12.080 


—35 


.04514 


25 


.494 




S 


—30 


13.900 


—30 


.05132 


50 


.501 




C 


—25 


15.940 


—25 


.05816 


75 


.508 




I 


—20 


18.220 


—20 


.06573 


100 


.515 




B 


-15 


20.760 


—15 


.07406 


125 


.523 




L 


—10 


23.590 


—10 


.08322 


150 


.530 




E 


—5 


26.730 


—5 


.09326 


175 


.538 









30.210 





.10420 


200 


.546 






5 


34.040 


5 


.11620 


225 


.554 






10 


38270 


10 


.12930 


250 


.562 






15 


42.920 


15 


.14340 


275 


.571 






20 


48.020 


20 


.15880 


300 


.579 






25 


53.600 


25 


.17540 


325 


.588 






30 


59 690 


30 


.19340 


350 


.597 






35 


66.330 


35 


.21270 


375 


.606 






40 


73.549 


40 


.23350 


400 


.615 






45 


81.400 


45 


.25590 


425 


.625 






50 


89.900 


50 


.27980 


450 


.635 






55 


99099 


55 


.30550 


475 


.645 






60 


109 000 


60 


.33290 


500 


.655 






65 


119 700 


65 


.36210 


525 


.665 






70 


131.299 


70 .39320 


550 


.675 






75 


143.699 


75 .42630 


575 


.686 






80 


157.000 


80 .46150 


600 


.697 






85 


171.199 


85 .49870 







96 



BUTADIENE 



BDI 



Lquefed compressed Colorless 



Gaeoane fta odor 



Exposure 



Water 

Pollution 



Vapor may explode n «nrted in art enclosed area 



May be dangerous rl 



1. RESPONSE TO DISCHARGE 
(Bm Response Method* Handbook) 



3. CHEMICAL DESIGNATIONS 

3.1 CO CompaUbWty Dam Olefin 

3.2 formula: CHi = CHCH . Cm, 

3.3 NtQ/UN Designation: 2.0/1010 

3.4 DOT ID No.: 1010 

15 CAS Registry Noj 106-99-0 



2 LABEL 

Category: FtammabM liquid 



4 OBSERVABLE CHARACTERISTICS 



5. HEALTH HAZARDS 

8.1 Personal Protective Equipment Chemical-type salety goggles; r 

those entering a tank or enclosed storage space, hose mask w 
atmosphere: self-contained breathing apparatus, rubber suit 

6.2 Symptoms Following Exposure: Slight anesthetic effect at high 

"frostbite" from skin contact, slight rmat»n to eyes and nose i 

5.3 Treatment of Exposure: Remove from exposure immediately Call a physician inhalation if 

breathing ts irregular or stopped, start resuscitation, administer oxygen. SKIN CONTACT: remove 
a clothing and wash affected skin area EYE CONTACT *ngate with water for IS 



Threshold Umrt Value: t .000 ppm 
Short Term Inhalation Untits: Data not evaiiabit 
Toxicity by Ingestion: Data not available 
Late Toxicity None 

Vapor (Oaa) Irritant Characteristics; Vapors ca. 
Thee 



Liquid or Solid d 



t Che red eristics I 



II spilled on clothing a 



L FIRE HAZARDS 
xh Point 105"F (eat) 

l Limits * Am 20%-l1.5% 
fU Rre Ejrtingukshing Agent* Stop no* d 

*,* Rre Extingutahing Agents Not to be 



Not pertinent 
Swhevtor m Ftrst: Vapors heavier than air 
and may travel a considerable distance I 
a source of tgnrtion and flashback. 
Ccnuuners may exptooe in a fire due to 

Electrical Hazard: CktSS 1. Group B 
■urnlng Rata: 9.0 mm/mm. 

Data not available 



7. CHEMICAL REACTIVITY 



7 J ttebWry During Transport Exptoerve 



7.6 Poh/mert aa tto n. Stable when inheritors 



1-Butylcatechol (0.01-0.02%) 



8- WATER POLLUTION 
6.1 Aquatic Toxicity: Not pertinent 
8 a Waterfowl Toxicity: Not pertinent 
•.3 BJotogicaf Oxygen Dernand (BOD): 

Notperbnent 
a.4 Food Cham C on cen tra tion Potential: 



10. HAZARD ASSESSMENT CODE 
(See Hazard Aaaaaamarrt Haw 
A-B-C-D-E-F-G-Z 



lL HAZARD CLASSIFICATIONS 
11.1 Coda ol Federal Regulation*: 

Flammable gas 
11 J MAS Hazard Rating for Bulk Weler 

Category Rating 

Rre 4 

Health 

Vapor Irritant 1 

UquM) or Solid Irritant 1 

Water Poiution 

Aquatic Toxicity . 

Aesthete Effect 1 

Reactivity 

Other Chemicals 1 

Water 2 

Serf Reaction 

11 J NFPA Hit 

Category 

Hearth Hazard (Blue) 

FLammabilrty (Red) 

Reactivity (Yellow) 



9. SHIPPING INFORMATION 

9.1 Qredes of Purity: Research grade 99 6 
moie% Special purtty 99 5 mows, 
Rubber grade: Ccrnmeretal: 

J Storage Temperature: Amt-ent 

%A Inarl Atmoeprterr No raqulremeni 

9.4 venting: Safety relief 



PHYSICAL AND CHEMICAL PROPERTIES 
Physical State at 15'C and 1 etm: 

Molecular Weight Data not available 
BoWng Point at 1 atm: 

24.1'F - — 4.4*C = 268.8*K 

_164'F = — 106.9*C - 164.3*K 
CritJcal Temperature: 

306'F = 152'C = 425-K 
Critical Pressure: 

628 psta = 42.7 atm = 4 32 MN/m> 
SpechSc Qravrty: 

621 at 20'C (liquid) 
Liquid Surface Tension: 

13.4 dynes/cm = 0134 N/m al 22'C 
2.6 Liquid Water Interracial Tension: 



Latent Heat of Vaporization: 

160 Btu/lb = 100 cal/g = 

4.19 X 10* J/kg 
Heat Of Combustion: — 19,009 Btu/lb - 

—10.560 cal/g = —442.13 X 10» J/kg 
Heat of Decomposition: Not pertinent 
Heat of Solution: Not pertinent 
Heat of Potymarfxation: —649 Btu'ib 

= —305 cal/g = -12.8 X 10* J/kg 
Heal Of Fusion: 35.28 cal/g 
Limiting Value: Data not available 
field Vapor Pressure: 6 1 psia 



6. FIRE HAZARDS (Continued) 



97 




12.17 


12.18 


12.19 




SATURATED LIQUID DENSITY 


UQUID HEAT CAPACITY 


LIQUID THERMAL CONDUCTIVITY 


UQUID VISCOSITY 


Temperature 


Pounds per cubic 


Temperature 


British thermal unit 


Temperature 


British thermal 






(degrees F) 


foot 


(degrees F) 


per pound-F 


(degrees F) 


square loot-F 


(degrees F) 


CenUpoise 


—110 


45.610 


—110 


.453 




N 


—110 


.437 


— 100 


45.220 


— 100 


.459 







—100 


.404 


—90 


44.840 


—90 


.465 




T 


—90 


.375 


—80 


44.460 


—80 


.471 






—80 


.349 


—70 


44.080 


—70 


.478 




P 


—70 


.326 


—60 


43.700 


—60 


.484 




E 


—60 


.306 


-50 


43.320 


—50 


.490 




R 


—50 


.288 


—40 


42.940 


—40 


.496 




T 


—40 


.272 


—30 


42.550 


—30 


.502 




I 


—30 


.258 


—20 


42.170 


—20 


.508 




N 


—20 


.245 


-10 


41.790 


— 10 


.514 




E 


—10 


.233 





41.410 





.520 




N 





.222 


10 


41.030 


10 


.526 




T 


10 


.212 


20 


40.650 


20 


.533 






20 


.203 



12.21 
SOLUBILITY IN WATER 


12.22 
SATURATED VAPOR PRESSURE 


12.23 
SATURATED VAPOR DENSITY 


12.24 
IDEAL GAS HEAT CAPACITY 


Temperature 
(degrees F) 


Pounds per 100 
pounds of water 


Temperature 
(degrees F) 


Pounds per square 
inch 


Temperature 
(degrees F) 


Pounds per cubic 
foot 


Temperature 
(degrees F) 


British thermal unit 
per pound-F 




I 


-55 


1.795 


—55 


.02235 





.307 




N 


—50 


2.109 


—50 


.02594 


25 


322 




S 


—45 


2.466 


—45 


.02997 


50 


.336 




O 


—to 


2.872 


—40 


.03448 


75 


.350 




L 


—35 


3.331 


—35 


.03952 


100 


.364 




U 


—30 


3.847 


—30 


.04512 


125 


.377 




B 


—25 


4.426 


-25 


.05131 


150 


ion 




L 


—20 


5.074 


—20 


.05815 


175 


.403 




E 


—15 


5.796 


—15 


.06568 


200 


.416 






—10 


6.598 


—10 


.07394 


225 


428 






—5 


7.487 


—5 


.08297 


250 


.440 









8.468 





.09283 


275 


.451 






5 


9.549 


5 


.10350 


300 


.463 






10 


10.740 


10 


.11520 


325 


.474 






15 


12.040 


15 


12780 


350 


.485 






20 


13.460 


20 


.14140 


375 


495 






25 


15.010 


25 


.15600 


400 


.505 






30 


16.690 


30 


.17180 


425 


.515 






35 


18.520 


35 


.18870 


450 


.525 






40 


20.500 


40 


.20680 


475 


.535 






45 


22.650 


45 


.22610 


500 


.544 






50 


24.960 


50 


.24670 


525 


.553 






55 


27.440 


55 


.26870 


550 


.562 






60 


30.120 


60 


.29200 


575 


.571 






65 


32.980 


65 


.31680 


600 


.579 






70 


36.050 


70 


.34300 







98 



BUTANE 



BUT 



■— - 


Uquefied oompreaaed Colorteaa Gaerjans-lite odor 
Floats and boa* on water nammabte. vtaM© vapor doud n formed. 


^^3§~ n — 


Fire 


FLAMMABLE. 

Flashback along vapor trail may occur 

Vapor may exptodeJfjpnrted m an enclosed area 

S;:: • r* :' ga- >' pews-* 

Co= »*Kitz eoiiaine's a"B P'OUfC! me-. e nect<ng shui-of* win waie- 


Exposure 


VAPOR 

If Inhaled. wwl cauae rjozmeea or difficult breathing 
Not irritating to eye*, noae or throat 

• r-f.^j ne* s-Mj*: g-r-a-' i '«?.■»:.&*> 

uoun 

Wdi cauae frostbrta. 


Water 

Pollution 


Not harmful to aquatic He. 

May be dangerous If it entera water rnlakee 


1 RESPONSE TO DISCHARGE 
(See Rmpoam Metnode Handbook) 
Issue wimin^nijri ftanvnabtirty 
Reatnct acceu 
Evacuate area 


2. LABEL 
2.1 Category: Flammable gaa 
2-2 Oaaa:2 


3. CHEMICAL DESIGNATIONS 
1*1 CO CompaMbMHy Ctaaa. Paraffin 
3.2 Fofmuat: n-C«Hia 
J.J lUO/UN Designation: 2.0/1011 
14 DOT ID No. 1011 
3.5 CAS Registry Ho.: 106-974 


4. 06SERVABU CHARACTERISTICS 

Compreaaed gaa 
4 J Color Colorteaa 
4.1 Odor l*o gasoline 


S. HEALTH HAZARDS 
6.1 Personal Protactfve Equipment Seil-comamed breathing apparatus and safety goggle* 
B.2 Symptoms Foaowtng ExpcMure: Hrgh axpoaura produces drowsiness but no other evtfence ot 

5.3 Treatment of Expoeore. ORAL AND ASPIRATION: No treatment requn>d INHALATION: Guard 

against setf-in|ury if stuporoua. confused, or eneethetQed Apply artificial raaprabon if not 
breathing. Avord edmmistretjon ot epinephrine or other syrnpatrttrrwrtetic amme*. Pravanl 
aapnatxKTt of vomitua by proper poerbonrnc of the head Give symptomatic and aupportive 

6.4 TVaohoW Umfl Value: 600 ppm 

5.6 Short Term Inhaiatton UrnJta.- Data not available 
5.0 Toxicity by Ingestion: No) pertinent 

6.7 Lata Toxicity: None 

6 6 Vapor (Cut) trrnanl Characteristic*: None 

6.9 LkKrtd of Soiw irritant Charactenattc* No appreciable hazard Practcatly harmieu to the ahtn 

because rt ta vary volatile and evaporate* quckly from the akin Some froetbfle posaaXe 

6.10 Odor Threshold: 6 16 ppm 

6.11 IDLM Value: Dam not available 



I FIRE HAZARDS 

►h Point — 100'F (est) 

► umtts m am-. i,e%-e.4% 

rVe Exdngurahlng AoerrtS- Stop flow o 
gas 

Ftre Extiogulehmg Agents Not to be 

Ueacc Not pertinertt 
Special Hazard* of Combustion 



Behavior m Fire- Not pertinent 

IgnMon Temperature: 550'F 

Eiectrtcei Hazard: Claaa 1. Group D 

Burning Rate: 7 9 mm/mm 

a^aabetfc Flame Temperature: 2435. (EaL) 

ttotcMometrtc Air to Fuel Ratio: 

15.35 (Est.) 
Flame Temperature-. Data not available 



7. CHEMICAL REACTIVITY 



7 J Stsoarty During Transport Stable 
7 A N^rtra*dr*g Agents for Adda and 

Caustic* Not pertinent 
7.5 Poh/mertzaborv Not pertment 

Not pertment 



fl WATER POLLUTION 
6.1 Aquatic Toxicity: None 
e.2 Waterfowl Toxicity: None 

6.3 Biological Oxygen Demand (BOO): 

None 

8.4 Food Cham Con c en tr a ti on Potantlet 



9. SHIPPING INFORMATION 

6.1 Oradee of Purity: Research 99 .95%. 

Purs: 994%. Technical 97,6% 
9 J Storage Temperature: Amtoent 
9 J Inert Atmosphere: No n 
9.4 Verrbng: Safely n 



10. HAZARD ASSESSMENT CODE 

(See Hazard Aaaeaameni Handbook) 

A-B-C-D-E-F-G 



U HAZARD CLASSIFICATIONS 
Code ot Federal Rogulattona: 

Flammable gas 
HAS Hazard Rating for Buft Water 

Transportation: 

Category Rett 

Fire 4 



Vapor Irritant, 
Liquid or Solid li 



Water Polubon 

Aquatic Toxicity 

Aesthetic Effect 

Reactivity 
Other Chemicals ... 

Water .... .0 

Self Reaction 

HFPA Hazard Classification: 

Category Cleaetflcatlon 

Health Hazard (Blue) 1 

Flammabilrty (Red) 4 

Reactivity (Yellow) 



II PHYSICAL AND CHEMICAL PROPERTIES 
2.1 Physical State at 1ST and 1 atm 



12.2 


Moreeuisr Weight 58 12 


12.3 


Boiling Point st 1 atm: 




31,1'F - — 0.40-C = 272.72*K 


12.4 


Freezing Point 




— 216*F - — 138'C = 135'K 


12.5 


Critical Temperature: 




306'F - 152-C = 425'K 


12.6 


Critical Pressure: 




550 8 psia = 37 47 atm - 3.796 


12-7 


Specmc Ofavtty: 




60 at OX (liquid) 


12.6 


liquid Surface Tenaion: 




14 7 dynes/cm = 0147 N/m at 0"C 


12.9 


Uqutd Water Interfadal Tension: (esi ) 




65 dynes/cm ■ 0.065 N/m at 22'C 


12.10 


Vapor (Oaa) Specmc Gravity: 20 st 20' 


12.11 


Ratio of Specific Heats of Vapor (Gas 




1 092 


12.12 


Latent Heat of Vaporization; 




170 Btu/lb = 92 cai/g ■ 




3 9 X 10* J/kg 


12.13 


Heat ot Combustion: —19.512 Btu/lb = 




-10.640 cal/g - —453 65 X 10 B J/1 


12.14 


Heat of OacompoHnon: Not pertinent 


12.15 


Heal of Solution: Not pertinent 


12.19 


Heal Of Polymerization: Noi pertinent 


12-25 


Heat of Fualon: 19 16 ca'/g 


12.2* 


UmWng Value: Data not available 


12.27 


ReM Vapor Preaeure: 52 4 psia 



99 




12.17 


12.18 


12.18 




SATURATED LIQUID DENSITY 


LIQUID HEAT CAPACITY 


LIQUID THERMAL CONDUCTIVITY 


LIQUID VISCOSITY 


Temperature 


Pounds per cubic 


Temperature 


British thermal unit 


Temperature 


British thermal 
unit-Inch per hour- 
square foot-F 






(degrees F) 


foot 


(degrees F) 


per pound-F 


(degrees F) 


(degrees F) 


Cent] poise 


— 110 


41.940 


—30 


.535 




N 


—110 


.535 


—100 


41.630 


—20 


.542 




O 


—105 


.511 


—90 


41.320 


—10 


.550 




T 


—100 


.469 


-80 


41.000 





.557 






—95 


.468 


—70 


40.690 


10 


.564 




P 


—90 


.449 


-60 


40.380 


20 


.571 




E 


—85 


.431 


—50 


40.070 


30 


.578 




R 


-80 


.414 


—40 


39.750 








T 


—75 


.398 


—30 


39 440 








1 


—70 


.383 


—20 


39 130 








N 


—65 


.369 


—10 


38.820 








E 


—60 


.356 





38.510 








N 


-55 


.344 


10 


38.190 








T 


—50 


.332 


20 


37.880 










—45 


.321 


30 


37.570 










—40 

—35 

—30 

—25 

-20 

-15 

—10 

—5 



5 

10 

15 


.311 
.301 
.292 
.263 
.275 
.267 
.260 
.253 
.246 
.239 
.233 
.227 



12.21 
SOLUBILITY IN WATER 


12.22 
SATURATED VAPOR PRESSURE 


12J3 
SATURATED VAPOR DENSITY 


12.24 
IDEAL GAS HEAT CAPACITY 


Temperature 
(degrees F) 


Pounds per 100 
pounds ol water 


Temperature 
(degrees F) 


Pounds per square 
inch 


Temperature 
(degrees F) 


Pounds per cubic 
loot 


Temperature 
(degrees F) 


British thermal unit 
per pound-F 




I 


—90 


.420 


—90 


.00616 





.360 




N 


—80 


.624 


—80 


.00890 


25 


.377 




S 


-70 


.905 


—70 


.01257 


50 


.392 




O 


—60 


1.283 


—60 


.01739 


75 


.408 




L 


—50 


1.784 


—50 


.02358 


100 


.424 




U 


—40 


2.435 


—40 


.03142 


125 


.439 




B 


—30 


3.269 


—30 


.04119 


150 


.454 




L 


—20 


4.320 


—20 


.05320 


175 


.468 




E 


—10 


5.629 


—10 


.06778 


200 


.483 









7.237 





.08525 


225 


.497 






10 


9.192 


10 


.10600 


250 


.511 






20 


11.540 


20 


.13030 


275 


.525 






30 


14.340 


30 


.15860 


300 


.539 






40 


17.640 


40 


.19120 


325 


.552 






50 


21.510 


50 


.22850 


350 


.565 






60 


25.990 


60 


.27080 


375 


.578 






70 


31.160 


70 


.31850 


400 


.591 






80 


37.080 


80 


.37200 


425 
450 
475 
500 
525 
550 
575 
600 


.603 
.615 
.628 
.639 
.651 
.663 
.674 
.685 



100 



STYRENE 



STY 



Ptwwletnyten* 



Cotorleaa to egnt B — •! pt—Wn l c 



FWmtfili, rrftabng vapor » product 






Exposure 



Water 
Pollution 



Vapor may explode 1 



VAPOR 

Irritating to eyea, noeej 



■•W. g>.« oxyge" 



May be dangerous n 



1. RESPONSE TO DISCHARGE 
iS*- Reepone* Methode Hawttbooe.) 



3 CNEMICAL DESIGNATIONS 

t Olefin 



3.3 I MO UN D**>gnj Hon 3 H 2055 

3.4 DOT 10 No- 2055 

SJ CAS Regrttry No.; 10CM2-5 



2. LABEL 
11 Category- Flammable ■ 



4 OBSERVABLf CHARACTERISTICS 



4J Cotor. Cotortea* 
4J Odor Sweet a( low ooncantrabons, 
cnarectensbc pungent, sharp. 



S- HEALTH HAZAROS 
Personal Protecttve Equipment A*-*upp*ed mas* or approved canater. rubber or plesbc ( 






Treatment of Expo*** INHALATION nmow 
reeprabon i 'W>w INGESTION do NOT . 
SKIN Ofl E>E CONTACT fluafi with plenty C 






100 p 



r 30r 



Snort Term li 

Toxtcrty by mgeetton. Grade 2. LD»c - 5 to 5 g/kg 

Lata Toxicity Data not avaaabW 

Vapor (Oas) Irritant Cnaracteriettc*: Vapor? cause moderate a 

find r*gn concentrations unpleasant Th* afiact » temporary 
Liquid or SoMd Irrftant CK»ract»nat>c»- Causes amarong o< the alun and frst-degr** bums □ 



«. FIRE HAZARDS 




0. HAZARD ASSESSMENT CODE 


6. 1 Ram Point 93*F O C , S8*F C.C 


fS*e Hazard Aaaaaamant Handbook) 


*J FavnmabtaUmrtaln AJr i 1%4.1% 




A-T-U-2 


*o Fa-* EiOnoUshinc, Agerrta: WaWr log. 






team, carton doxxte, o dry ehemeaJ 






14 Fa-e EjrtnguJaNng Agerrta Not to be 
Uaad: Water may be rteftectrve 

0-5 Specs*) Hazard* of Cornbuebon 
Products: Not pertinent 








11 HAZARD CLASSIFICATIONS 
Code of Federal Regulation*; 


*.« Banarvtor kn FW: Vapor a heavier man air 
and may travel ocraaderabie distance to a 
eourc* of ignRion and flajh back Ai 
elevated temperature* such as m fre 


114 


Flammable kqud 
MAS Haxard Rabng tor fMk Watar 


oondrbons. poNmaraabon may take place 




^^ 3 


irw may lead to container explosion 

eJ Electrical Hazard: Class I. Group D 
6.S Durnmg Rat*: 5 : mm/nwi 




"^"^ 








Data not evaaabie 

(Continued) 










7. CHEMICAL REACnvrTY 






7.1 RaactMty With Watar: No raacoon 
7 J ReactMty arm Common laetartaeB: No 




Other CtwrracaJa 2 

Watar 


7 J StaMfty During Tranepcrt Stabla 
7.4 Nautraattng Aganta for Add* and 

C*uettc* Not porbnent 
7 £ Pon/rrtarizattort May occur rf heated 
above ISO'F Can cause rupture ot 


11J 


Heart* Hazard iBw 2 






container MeUI aaJU. peroxides, and 






strong aods may alao cauae 






porymeroabon 






7A Inhibitor of Porym*r&*boa Terbary 






butytcatechol. 10-15 pom 






7,7 Meter Ratio fflaactanl to 

Product): Data not available 
7 J Reecttvtty Group: 30 






12. 


PHYSICAL AND CHEMICAL PROPERTIES 




12.1 


Ptryaacai Stata at 1SX and i atnr 

Liquid 




IU 


laoaecular Wetottt 104 15 




12J 


293 4'F m 145-2*C - 41B.4*K 




HI 


Fraaxtng Feint 

— 23,1*F - — 30 6*C - 242.6-K 


a. WATER POLLUTION 


li Aquatic Toxicity- 




703*F - 373*C - fMS'K 


22 ppm/06 hr/biuegrll/TL,' fresh water 


UJ 




L2 Waterfowl Toxicity- Data not available 




580 paia - 39 *6 Mm - 4 00 MN/m' 


i.J EWc4ogtc»i Oryg*n Oarnand (BOOr. 


12.7 


Specmc Gvavtty: 






0.906 at 20'C (aqurj) 


1.4 Food Cham Co»uaoti afJon PWertbafc 


UJ 


Ltguw Surface TbjMfOR 


None 




32 14 dynee/cm - 003214 N/m at 
IB'C 
Liquid Watar amVarfacaal Tanaaort 




Utl 






35 48 dynaa/cm - 0.O354S N/m al 






19*C 




12.10 


Vapor (Oee) Speclftc Oravtty: 

Not pertinent 




12.11 


Rabo of apacmc Heats of Vapor t O*e t 
1.074 




12.12 


Latarrt Heal of Vaporixaflon: 




12.13 


3.63 X 10* J/hg 
Heat of Combuabon: Not penment 


9. SNIPPING INFORMATION 


l.i Oradee of Purity: 99 5 + % 


U.M 




92 Storage Temperature: Ambient 


1t.1l 


Heat of Sotwbon: Not partmem 


f J mart Atmc*prwr*: No r*our*men1 


ta.ii 


Ha*t of PorymwtiaUuiL —277 Btu/fe 


9.4 VerrOng; Open {nam* arrester) 




- —154 cai;g - -645 X 10« J/kg 




12-IS 


Heat of Fuaiort Data not avaaabte 




12_M 


Urnfdng Vakir Oata r«i avaaabia 




12J7 


Rett Vapor Praaaurr c :~ pm 


b FIRE HAZA 


RDS (Corttnuad) 


8i 1 Stoa^Momatnc Air to Fuel ftatto: Oau not i 


vaiaMe 




«-i: Faame Tamper a Ura: Oata not avadabie 







101 




12.17 
SATURATED LIQUID DENSITY 


12.16 
LIQUID HEAT CAPACITY 


12.10 
LIQUID THERMAL CONDUCTIVITY 


12.20 
UOUID VISCOSITY 


Temperature 
(degrees F) 


Pounds per cubic 
foot 


Temperature 
(degrees F) 


Britisri thermal unit 
per pound-F 


Temperature 
(degrees F) 


British thermal 
unit-inch per hour- 
square fcot-F 


Temperature 
(degrees F) 


Centi poise 


40 


57.430 





.389 


15 


1.087 


40 


.950 


50 


57.120 


5 


.391 


20 


1.080 


50 


.872 


60 


56.800 


10 


.393 


25 


1.074 


60 


.803 


70 


56.490 


15 


.395 


30 


1.067 


70 


.742 


80 


56.180 


20 


.397 


35 


1.060 


80 


.688 


90 


55.870 


25 


.399 


40 


1.054 


90 


.639 


100 


55.560 


30 


.401 


45 


1.047 


100 


.595 


110 


55.240 


35 


.403 


50 


1.040 


110 


.556 


120 


54.930 


40 


.405 


55 


1.033 


120 


.521 


130 


54.620 


45 


.407 


60 


1.027 


130 


.488 


140 


54310 


50 


.409 


65 


1.020 


140 


.459 


150 


54.000 


55 


.411 


70 


1.013 


150 


.433 


160 


53.680 


60 


.413 


75 


1.006 


160 


.408 


170 


53.370 


65 


.415 


80 


1.000 


170 


.386 


180 


53.060 


70 


.417 


85 


.993 


180 


.366 


190 


52.750 


75 


.419 


90 


.986 


190 


.347 


200 


52430 


80 


.421 


95 


.980 


200 


.330 


210 


52.120 


85 


.423 


100 


.973 


210 


.314 






90 


.424 


105 


.966 










95 


.426 


110 


.959 










100 


.428 


115 


.953 










105 


.430 


120 


.946 










110 


.432 














115 


.434 














120 


.436 











12.21 


12.22 


12.23 


12.24 


SOLUBILITY IN WATER 


SATURATED VAPOR PRESSURE 


SATURATED VAPOR DENSITY 


IDEAL GAS HEAT CAPACITY 


Temperature ' Pounds per 100 


Temperature 


Pounds per square 


Temperature 


Pounds per cubic 


Temperature 


British thermal unit 


(degrees F) pounds of water 


(degrees F) 


inch 


(degrees F) 


foot 


(degrees F) 


per pound-F 


68.02 .300 


40 


.034 


40 


.00066 





.239 




50 


.049 


50 


.00094 


25 


.253 


I 


60 


.070 


60 


.00131 


50 


.266 




70 


.099 


70 


.00181 


75 


.279 




80 


.137 


80 


.00247 


100 


.292 




90 


.188 


90 


.00332 


125 


.304 




100 


.254 


100 


.00440 


150 


.317 




110 


.339 


110 


.00577 


175 


.329 


I 


120 


.447 


120 


.00748 


200 


.340 




130 


.583 


130 


.00959 


225 


.352 




140 


.753 


140 


.01218 


250 


.363 


| 


150 


.963 


150 


.01532 


275 


.374 




160 


1.221 


160 


.01911 


300 


.385 




170 


1.534 


170 


.02364 


325 


.396 




180 


1.912 


180 


.02900 


350 


.406 


I 


190 


2.365 


190 


.03533 


375 


.416 


1 


200 


2.905 


200 


.04272 


400 


.426 


! 


210 


3.542 


210 


.05132 


425 


.435 




220 


4.292 


220 


.06126 


450 


445 




230 


5.167 


230 


.07269 


475 


.454 




240 


6.183 


240 


.08575 


500 


.462 




250 


7.358 


250 


.10060 


525 


.471 




260 


8.709 


260 


.11740 


550 


.479 




270 


10.250 


270 


.13630 


575 


.487 


1 


280 


12.010 


280 


.15760 


600 


.495 




290 


14.010 


290 


.18130 







"U.S. GOVERNMENT PRINTING OFF I CE : 1 99 0- 281-626 : 2001 



U581 



HE 199.5 .D3 

90/02 

Derailment of a CSX 

transportation freight 

train and fire involving 



National Emergency Training Center 

Learning Resource Center 

16825 South Seton Avenue 

Emmitsburg, MD 21727 

NETC LRC 



5^99 




002233 



rrx-yl OO 



"IONAL TRANSPORTATION SAFETY BOARD 
Washington, D.C. 20594 



Official Business 

PENALTY FOR PRIVATE USE, $300 



THIRD CLASS BULK RATE 

POSTAGE & FEES PAID 

NATIONAL TRANSPORTATION 

SAFETY BOARD 

PERMIT NO. G-200 



NTSB - 14A 
NAT"L EMERGENCY 

LIBRARY 

16825 SOUTH SETON 
ITS3URG, MO 



{ 1) 
TRAINING CNTR 

AVE 

21727